Chain quality monitoring information visualization method and system based on meteorological observation engineering

Abstract

The application relates to the technical field of big data visualization, and provides a chain quality monitoring information visualization method and system based on meteorological observation engineering. The method comprises the following steps: collecting engineering quality monitoring data of each key quality control node in the implementation of meteorological observation engineering; constructing a whole-process stage quality monitoring model comprising a node quality feature description unit, a quality influence transmission path unit between nodes and a quality control threshold interval unit according to the engineering quality monitoring data; then, combining quality control permission information of each participant in a matrix organization collaborative framework, performing correlation analysis on the related units of the model, and generating a chain collaborative interaction strategy; finally, converting the engineering quality monitoring data into multidimensional dynamic views comprising a quality control node correlation view, a quality influence transmission path view and a quality index change trend view according to the strategy, performing abnormal mode identification on the quality index change trend view, and generating hierarchical early warning information comprising early warning grades and early warning ranges.

Description

Technical Field

[0001] This application belongs to the field of big data visualization technology, specifically involving a chain-based quality monitoring information visualization method and system based on meteorological observation engineering. Background Technology

[0002] Meteorological observation projects, as a crucial means of acquiring meteorological information, directly impact the accuracy and reliability of weather forecasts. Quality monitoring is a key element in ensuring these projects achieve their intended goals. Traditional quality monitoring relies primarily on manual inspections and localized data monitoring. Data collection often involves manually recording partial data at key points periodically, a method that is inefficient and prone to data omissions and errors. Quality analysis typically focuses on evaluating individual stages or localized data, lacking in-depth analysis of the inherent relationships between various quality control points. In terms of collaborative management, meteorological observation projects involve multiple departments and levels, leading to poor communication and collaboration among stakeholders, resulting in disjointed processes. Regarding quality visualization, traditional methods only provide simple reports and static charts, failing to intuitively display the dynamic changes in project quality and the overall process. Finally, the lack of effective intelligent identification and tiered early warning mechanisms hinders the timely detection of potential quality issues.

[0003] In summary, traditional quality monitoring methods, due to high technical barriers and complex processes, result in fragmented quality monitoring, making it difficult to form a complete quality control system and failing to comprehensively and accurately reflect the quality status of the project. Summary of the Invention

[0004] This application provides a chain-based quality monitoring information visualization method and system based on meteorological observation engineering.

[0005] In a first aspect, embodiments of this application provide a chain-based quality monitoring information visualization method based on meteorological observation engineering, applied to a chain-based quality monitoring information visualization system, the method comprising:

[0006] Collect engineering quality monitoring data generated at each key quality control node during the implementation of the meteorological observation project;

[0007] Based on the engineering quality monitoring data, a full-process stage quality monitoring model reflecting the inherent correlation of each quality control node is constructed. The full-process stage quality monitoring model includes node quality characteristic description unit, quality influence transmission path unit between nodes, and quality control threshold interval unit.

[0008] By combining the quality control authority information of each participant in the matrix-style organizational collaborative architecture, the node quality feature description unit and the quality impact transmission path unit between nodes in the full-process stage quality monitoring model are correlated and analyzed to generate a chain-style collaborative interaction strategy for coordinating the quality control behavior of each participant.

[0009] The engineering quality monitoring data is converted into a multi-dimensional dynamic view according to the chain-based collaborative interaction strategy. Anomaly pattern recognition is performed on the quality indicator change trend view in the multi-dimensional dynamic view. Based on the recognition results, hierarchical early warning information including early warning level and early warning range is generated. The multi-dimensional dynamic view includes a quality control node association view, a quality impact transmission path view, and a quality indicator change trend view.

[0010] Secondly, embodiments of this application provide a chain-based quality monitoring information visualization system, which includes a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the above-described method.

[0011] Thirdly, embodiments of this application provide a computer-readable storage medium including a computer program, which, when run on a chain-based quality monitoring information visualization system, causes the chain-based quality monitoring information visualization system to perform the steps of the above-described method.

[0012] This application's embodiments start by collecting engineering quality monitoring data from key quality control nodes in meteorological observation projects, then constructing a full-process stage quality monitoring model, generating a chain-like collaborative interaction strategy, and finally realizing multi-dimensional dynamic view display of engineering quality monitoring data, anomaly pattern recognition, and hierarchical early warning. Overall, it solves the problems of fragmented quality monitoring and low collaborative efficiency caused by high technical barriers and poor process connection in meteorological observation engineering.

[0013] In detail, a full-chain, visualized quality control system built upon a chain-based dual-engine management approach digitizes key quality control points at each stage of the project, constructing a comprehensive quality monitoring indicator system. This transforms fragmented quality monitoring into a holistic system, comprehensively and systematically reflecting the project's quality status. Combined with a matrix-style organizational collaborative architecture, a cross-level, cross-departmental collaborative interaction mechanism, leveraging a chain-based collaborative interaction strategy, effectively improves the collaborative efficiency among all participants, breaks down departmental barriers, and ensures timely and accurate information transmission, guaranteeing collaborative work and jointly safeguarding project quality. A visualization model transforms quality data into multi-dimensional dynamic views, achieving a visualized presentation of quality status, process deviations, and rectification loops. This allows relevant personnel to intuitively understand the real-time status of project quality, promptly identify problems, and take corrective measures. Simultaneously, the designed quality early warning strategy intelligently identifies and classifies abnormal data based on preset thresholds, enabling timely detection of potential quality issues and tiered warnings according to the severity of the problems, improving the timeliness and effectiveness of quality control. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating a chain-based quality monitoring information visualization method based on meteorological observation engineering, provided in an embodiment of this application.

[0015] Figure 2 This is a schematic diagram of the structure of a chain-type quality monitoring information visualization system provided in an embodiment of this application. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this application. Obviously, the described embodiments are only some embodiments of the technical solutions of this application, and not all embodiments. Based on the embodiments recorded in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the technical solutions of this application.

[0017] See Figure 1 This application provides a chain-based quality monitoring information visualization method based on meteorological observation engineering. This method can be applied to a chain-based quality monitoring information visualization system. The specific process is as follows: steps 110-140.

[0018] Step 110: Collect engineering quality monitoring data generated at each key quality control node during the implementation of the meteorological observation project.

[0019] In the application scenario of this embodiment, the implementation of meteorological observation engineering involves multiple key quality control nodes, which generate a large amount of data related to engineering quality during the implementation process. For example, at the meteorological observation equipment installation node, data on equipment installation accuracy and installation environment parameters are generated; at the meteorological data acquisition node, data on the accuracy and completeness of the acquired data are generated. By deploying data acquisition equipment and systems at each key quality control node, this engineering quality monitoring data is collected in real time. The data acquisition equipment can be sensors, monitoring instruments, etc., capable of accurately acquiring various parameters related to engineering quality. This collected data will be used to build quality monitoring models, generate collaborative interaction strategies, and perform anomaly warnings.

[0020] Step 120: Based on the engineering quality monitoring data, construct a full-process stage quality monitoring model that reflects the inherent correlation of each quality control node. The full-process stage quality monitoring model includes node quality feature description units, quality influence transmission path units between nodes, and quality control threshold interval units.

[0021] Based on the collected engineering quality monitoring data, a full-process quality monitoring model reflecting the inherent relationships among various quality control nodes needs to be constructed. This model comprises three important units: a node quality characteristic description unit, a quality impact transmission path unit between nodes, and a quality control threshold interval unit. The node quality characteristic description unit describes the quality characteristics of each key quality control node. Through analysis of the engineering quality monitoring data, key attributes representing node quality are extracted and organized into a structured feature vector. The quality impact transmission path unit represents the transmission relationship of quality impact between different nodes. By analyzing the temporal correlation and causal relationship between nodes in the data, the direction and intensity of quality impact transmission are determined. The quality control threshold interval unit sets the allowable fluctuation range of quality attributes for each key quality control node at different implementation stages. This range is determined through statistical analysis of historical data and in conjunction with engineering quality acceptance standards.

[0022] Step 121: Identify the feature fields corresponding to each key quality control node in the engineering quality monitoring data and distinguish the differences in data features of different node types to generate a node data classification set.

[0023] When constructing a quality monitoring model for the entire process, the first step is to process the collected engineering quality monitoring data. This involves identifying the characteristic fields corresponding to each key quality control node in the data; these characteristic fields are important data items that reflect the quality of each node. For example, for the calibration node of meteorological observation equipment, characteristic fields might include calibration accuracy and calibration time. Different types of key quality control nodes have different data characteristics, which need to be distinguished. Through data analysis and comparison, data with similar characteristics are grouped together, generating a node data classification set.

[0024] Step 122: By analyzing the key attributes reflecting the quality status in the data of the node data classification set and the manifestation of the key attributes, generate quality attribute description information for each key quality control node, organize the quality attribute description information into a structured feature vector, and generate node quality feature description unit.

[0025] After obtaining the node data classification set, further analysis is performed on the key attributes reflecting the quality status. These key attributes are factors that directly reflect the quality of the nodes, such as the accuracy and stability of meteorological observation data. Simultaneously, the representation format of these key attributes needs to be determined, such as whether it is numerical, character, or Boolean. Based on the analysis results, quality attribute description information for each key quality control node is generated. This quality attribute description information is organized into structured feature vectors, with each feature vector representing a quality characteristic of a key quality control node. These feature vectors constitute the node quality feature description unit, which can accurately describe the quality status of each key quality control node.

[0026] Step 123: Perform time-series correlation analysis on the quality attribute description information of each key quality control node, capture the quality characteristic change sequence of different nodes in the time dimension, and identify potential correlation clues between nodes through sequence comparison.

[0027] To determine the quality impact relationships between different key quality control nodes, a time-series correlation analysis of the quality attribute description information of each node is necessary. Since project implementation is a continuous process, the quality characteristics of different nodes may influence each other over time. By analyzing the change sequences of quality attribute description information over time, patterns in the changes of quality characteristics between different nodes can be captured. Comparing these change sequences reveals potential correlation clues, such as whether a change in the quality characteristics of one node will cause changes in the quality characteristics of other nodes after a certain period. These potential correlation clues can be used to construct quality impact transmission path units between nodes.

[0028] Step 124: Based on the potential correlation clues and the logical dependencies of the engineering implementation process, analyze the direction of the quality influence between different key quality control nodes and construct the initial directed connection relationship between the nodes.

[0029] After identifying potential correlations between nodes regarding quality impacts, and considering the logical dependencies of the engineering implementation process, the direction of quality impacts between different key quality control nodes is further analyzed. Engineering implementation processes typically have a certain sequence and logical relationships; for example, the data acquisition node can only begin after the installation of meteorological observation equipment is completed. Based on these logical relationships and potential correlations, it is determined from which node the quality impact is transmitted to which node, thus constructing initial directed connections between nodes. These directed connections represent the direction of quality impact transmission between different nodes and are used to construct a directed weighted graph structure.

[0030] Step 125: By analyzing the causal correlation of node feature changes in historical data, determine the influence degree parameter of each directed connection, and perform weighted processing on the initial directed connection relationship based on the influence degree parameter to construct a directed weighted graph structure. The directed weighted graph structure is determined as the quality influence transmission path unit between nodes.

[0031] To more accurately describe the transmission relationship of quality influence between nodes, it is necessary to determine the influence degree parameter for each directed connection. By analyzing the causal correlation of node characteristic changes in historical data, the strength of quality influence between different nodes can be identified. For example, whether the quality change of one node has a large or small impact on the quality of another node can be determined by statistically analyzing the correlation of quality characteristic changes between the two nodes in historical data. Based on the determined influence degree parameter, the initial directed connection relationships are weighted. Each directed connection is assigned a weight value, which represents the strength of the quality influence on that connection. After weighting, a directed weighted graph structure is constructed, which can more accurately represent the transmission path of quality influence between nodes. This directed weighted graph structure is determined as the unit of the quality influence transmission path between nodes.

[0032] Step 126: Screen historical quality compliance samples from the engineering quality monitoring data, perform statistical characteristic analysis on the quality attribute data of each key quality control node in the historical quality compliance samples, and extract the distribution pattern characteristics and dispersion index of quality attribute data under different implementation stages.

[0033] To determine the quality control threshold interval, it is necessary to screen historical quality compliance samples from the engineering quality monitoring data. Historical quality compliance samples refer to sample data where the quality of each key quality control node in the past project implementation process met the engineering quality acceptance standards. Statistical characteristic analysis is performed on the quality attribute data of each key quality control node in these historical quality compliance samples, such as calculating the mean, median, and standard deviation of the data. Through these statistical analyses, the distribution characteristics of the quality attribute data at different implementation stages are extracted, such as whether the data follows a normal, skewed, or other distribution pattern. Simultaneously, dispersion indicators, such as standard deviation and variance, are extracted; these indicators reflect the data fluctuation.

[0034] Step 127: Based on the distribution morphology characteristics and the dispersion index, and in conjunction with the engineering quality acceptance standards, determine the allowable fluctuation range of quality attributes for each key quality control node at different implementation stages, and define the allowable fluctuation range of quality attributes as the quality control threshold interval unit.

[0035] Based on the extracted distribution characteristics and dispersion indicators, and in conjunction with engineering quality acceptance standards, the permissible fluctuation range of quality attributes for each key quality control node at different implementation stages is determined. The engineering quality acceptance standards specify the quality requirements for each key quality control node at different stages. Based on these requirements and the statistical characteristics of the data, a reasonable fluctuation range is determined. For example, if the quality attribute data of one of the key quality control nodes follows a normal distribution and has a small standard deviation, the permissible fluctuation range can be relatively narrow; if the data has a large dispersion, the permissible fluctuation range can be appropriately widened. The determined permissible fluctuation range of quality attributes is used as the quality control threshold interval unit for quality monitoring and anomaly early warning.

[0036] Step 128: After completing the multi-unit collaborative verification of the node quality feature description unit, the inter-node quality influence transmission path unit, and the quality control threshold interval unit, the node quality feature description unit, the inter-node quality influence transmission path unit, and the quality control threshold interval unit are integrated to generate a full-process stage quality monitoring model that reflects the inherent correlation of each quality control node.

[0037] After constructing the node quality characteristic description unit, the inter-node quality impact transmission path unit, and the quality control threshold interval unit, a multi-unit collaborative verification is required. This involves checking the data consistency and logical rationality among these three units to ensure they cooperate effectively and accurately reflect the inherent relationships between each quality control node. For example, it checks whether the quality impact transmission direction and intensity determined in the inter-node quality impact transmission path unit match the data in the node quality characteristic description unit, and whether the allowable fluctuation range set in the quality control threshold interval unit is reasonable. After completing the collaborative verification, these three units are integrated to generate a complete end-to-end quality monitoring model. This model comprehensively reflects the inherent relationships between each quality control node during the implementation of the meteorological observation project.

[0038] Step 130: Combining the quality control authority information of each participant in the matrix-style organizational collaborative architecture, perform correlation analysis on the node quality feature description unit and the quality impact transmission path unit between nodes in the full-process stage quality monitoring model, and generate a chain-style collaborative interaction strategy for coordinating the quality control behavior of each participant.

[0039] In the implementation of meteorological observation projects, a matrix-style collaborative organizational structure is typically adopted, involving multiple stakeholders such as equipment suppliers, construction contractors, and quality supervisors. Each stakeholder has different quality control permissions. Based on the quality control permission information of these stakeholders, a correlation analysis is performed on the node quality characteristic description units and the quality impact transmission path units between nodes in the full-process quality monitoring model. The analysis examines the characteristic access permissions of each stakeholder to different key quality control nodes and their intervention permissions on the quality impact transmission path, identifying correlations and conflicts between permissions. Through this correlation analysis, a chain-like collaborative interaction strategy is generated to coordinate the quality control behaviors of all stakeholders. This strategy defines the responsibilities and operating procedures of each stakeholder in the quality control process, ensuring that all stakeholders can work collaboratively to jointly guarantee project quality.

[0040] Step 131: Perform permission scope division processing on the quality control permission information of each participant in the matrix-style organizational collaborative architecture to obtain a permission description set that includes node control permission items and path intervention permission items.

[0041] To better perform correlation analysis and generate chain-like collaborative interaction strategies, it is necessary to process the quality control permission information of each participant in the matrix organizational collaborative architecture. The permission information is divided into node control permission items and path intervention permission items. Node control permission items represent each participant's access and operation permissions for different key quality control nodes, such as whether a participant has the right to view the quality characteristic data of a node or adjust the quality of a node. Path intervention permission items represent each participant's intervention permissions for different quality impact transmission paths, such as whether they can change the transmission direction of the path or adjust the impact intensity of the path. These permission items are organized and integrated to obtain a permission description set containing node control permission items and path intervention permission items. This set is used for mapping matching and correlation analysis.

[0042] Step 132: Perform mapping and matching processing between the node control permission items in the permission description set and the node quality feature description unit to determine the feature access permissions of each participant to different key quality control nodes, and generate a node permission mapping relationship table.

[0043] After obtaining the set of permission descriptions, the node control permission items within them are mapped and matched with the node quality feature description units in the full-process stage quality monitoring model. By comparing the feature fields in the node control permission items and the node quality feature description units, the feature access permissions of each participant to different key quality control nodes are determined. For example, if one participant's node control permission item includes access to the calibration accuracy feature of the meteorological observation equipment calibration node, the corresponding calibration accuracy feature field is found in the node quality feature description unit, determining that the participant has the right to access the data in that field. The feature access permissions of each participant to different key quality control nodes are organized and recorded to generate a node permission mapping table, which comprehensively represents the permission relationships of each participant in accessing node quality features.

[0044] Step 133: Perform association matching processing on the path intervention permission items in the permission description set and the quality impact transmission path unit between nodes to determine the intervention permission level of each participant on different quality impact transmission paths and generate a path permission association matrix.

[0045] Similarly, the path intervention permission items in the permission description set are associated and matched with the quality impact transmission path units between nodes. The intervention permissions of each participant for different quality impact transmission paths are analyzed in the path intervention permission items, and the intervention permission level of each participant for different paths is determined based on these permissions. For example, one participant may have high-level intervention permissions for a certain quality impact transmission path, while another participant may only have low-level intervention permissions. The intervention permission levels of each participant for different quality impact transmission paths are organized and recorded to generate a path permission association matrix. This matrix can comprehensively characterize the permission relationships of each participant in terms of quality impact transmission path intervention.

[0046] Step 134: Based on the node permission mapping table and path permission association matrix, perform permission conflict analysis and processing, identify the permission overlap area of ​​different participants on the same node or path, and generate permission conflict identification results.

[0047] After generating the node permission mapping table and the path permission association matrix, permission conflict analysis needs to be performed on these two tables. Different participants may have overlapping permissions when performing quality control on the same node or path, leading to management confusion and conflicts. By analyzing the node permission mapping table and the path permission association matrix, overlapping permission areas for different participants on the same node or path can be identified. For example, two participants may both have permission to modify features of a key quality control node, or both may have advanced intervention permissions for a quality impact propagation path. These overlapping permission areas are recorded, and the type, location, and involved participants of the conflict are analyzed to generate permission conflict identification results.

[0048] Step 1341: Traverse the node permission mapping table, extract multiple participants and their node control permission items corresponding to each key quality control node, compare the range intersection of the node control permission items of different participants, and mark the node permission conflict area when the range intersection is not empty.

[0049] When performing permission conflict analysis, the node permission mapping table is first traversed. For each critical quality control node, multiple participating parties and their corresponding node control permission items are extracted. The intersection of the ranges of these different participating parties' node control permission items is compared. If the intersection is not empty, it indicates that there is permission overlap, and this area is marked as a node permission conflict area. For example, for the meteorological observation equipment installation node, participating party A's node control permission items include access and modification permissions for installation accuracy and installation time, and participating party B's node control permission items also include access and modification permissions for installation accuracy. In this case, the intersection of the installation accuracy permission items is not empty, and it is marked as a node permission conflict area.

[0050] Step 1342: Traverse the path permission association matrix, extract multiple participants and their path intervention permission levels corresponding to each quality impact transmission path, compare the relationship between the path intervention permission levels of different participants, and mark the path permission conflict area when there is overlap of same or higher level permissions.

[0051] Next, the path permission association matrix is ​​traversed. For each quality impact transmission path, multiple participants and their corresponding path intervention permission levels are extracted. The relationship between the path intervention permission levels of different participants is compared. If there is overlap between the same or higher-level permissions, the area is marked as a path permission conflict area. For example, for a certain quality impact transmission path, if both participant C and participant D have high-level intervention permissions, then this path has a permission conflict and is marked as a path permission conflict area.

[0052] Step 1343: Collect all marked node permission conflict areas and path permission conflict areas, record the participant identifiers, conflict node or path identifiers and conflict permission content involved in each conflict area, and generate permission conflict identification results containing conflict type, conflict location and conflict participants.

[0053] After marking all node and path permission conflict areas, information on these conflict areas is collected. The identifiers of the participants involved in each conflict area are recorded to identify which participants are in conflict. The conflicting node or path identifiers are recorded to determine the specific location where the conflict occurs. Simultaneously, the conflicting permission content is recorded to identify which permissions overlap and conflict. This information is then organized and summarized to generate a permission conflict identification result containing the conflict type, conflict location, and conflicting participants. This result can be used for permission coordination processing.

[0054] Step 135: Based on the permission conflict identification results, perform permission coordination processing on the node permission mapping relationship table and the path permission association matrix to generate a chain-like collaborative interaction strategy for coordinating the quality control behaviors of each participant.

[0055] Based on the results of permission conflict identification, permission coordination processing is required for the node permission mapping table and path permission association matrix. By adjusting the permissions of each participant, permission conflicts are resolved, ensuring that all participants can work collaboratively. Based on the adjusted node permission mapping table and path permission association matrix, a new chain-like collaborative interaction strategy is generated to coordinate the quality control behaviors of each participant. This strategy will define the specific responsibilities and operating procedures of each participant in the quality control process, avoiding management chaos and quality problems caused by permission conflicts.

[0056] Step 1351: Prioritize the node permission conflict areas in the permission conflict identification results, determine the permission priority order according to the hierarchical relationship of the participants in the matrix organizational collaborative architecture, retain the node control permission items of the first priority participants as primary permissions, and adjust the node control permission items of the second priority participants to auxiliary permissions.

[0057] When handling permission coordination, conflicting permission areas in the permission conflict identification results are prioritized. The priority order is determined based on the hierarchical relationship of the participants within the matrix organizational structure. For example, in a matrix organizational structure, the quality supervisor might be at a higher level than the construction contractor, thus the quality supervisor has higher permission priority. The node control permissions of the first-priority participants are retained as primary permissions, which can be used for comprehensive management and control of node quality. The node control permissions of the second-priority participants are adjusted to secondary permissions, which can be used for a certain degree of supervision and feedback on node quality. This approach resolves node permission conflicts and ensures the orderly implementation of node quality control.

[0058] Step 1352: Perform intervention scope division processing on the path permission conflict area in the permission conflict identification result. Divide the path into core path segments and non-core path segments according to the importance parameter of the quality impact transmission path. Assign the intervention permission of the core path segment to the participants with the target path intervention permission level, and assign the intervention permission of the non-core path segment to other participants.

[0059] For the path permission conflict areas identified in the permission conflict identification results, intervention scope division is performed. Based on the importance parameter of the quality impact transmission path, the paths are divided into core path segments and non-core path segments. Core path segments have a significant impact on project quality, while non-core path segments have a relatively smaller impact. Intervention permissions for core path segments are assigned to stakeholders with the target path intervention permission level. These stakeholders typically possess high professional capabilities and management authority, enabling effective intervention and control of core path segments. Intervention permissions for non-core path segments are assigned to other stakeholders. This approach fully utilizes the resources of each stakeholder, improving the efficiency of quality control.

[0060] Step 1353: Construct a permission collaboration rule base based on the adjusted node permission mapping table and path permission association matrix. The permission collaboration rule base includes node permission allocation rules, path intervention collaboration rules, and conflict resolution triggering rules.

[0061] After coordinating node and path permissions, a permission collaboration rule base is constructed based on the adjusted node permission mapping table and path permission association matrix. This rule base contains three important rules: node permission allocation rules, path intervention collaboration rules, and conflict resolution triggering rules. The node permission allocation rules specify how each participant allocates permissions to different key quality control nodes, ensuring that quality control at each node has a clearly defined responsible party. The path intervention collaboration rules specify the collaboration methods and processes for each participant when intervening in the transmission path of quality impacts, avoiding quality problems caused by improper intervention. The conflict resolution triggering rules specify the resolution methods and processes when permission conflicts occur, ensuring that conflicts are resolved promptly.

[0062] Step 1354: Generate quality control behavior guidance information for each participant based on the permission collaboration rule base. The behavior guidance information includes node feature access process, path intervention operation specifications, and cross-participant collaborative response mechanism.

[0063] Based on the permission collaboration rule base, quality control behavior guidelines for each participant are generated. These guidelines comprise three key parts: node feature access procedures, path intervention operation specifications, and cross-partner collaborative response mechanisms. The node feature access procedures specify the concrete steps and requirements for each participant when accessing feature data of key quality control nodes, ensuring data security and accuracy. The path intervention operation specifications outline the specific operating methods and precautions for each participant when intervening in the quality impact transmission path, preventing quality issues caused by improper operation. The cross-partner collaborative response mechanism specifies the collaborative response methods and procedures for each participant when encountering quality issues, ensuring timely and effective problem-solving.

[0064] Step 1355: Combine the aforementioned permission collaboration rule base and quality control behavior guidance information to generate a chain-like collaborative interaction strategy for coordinating the quality control behaviors of all participating parties.

[0065] By combining a permission-based collaborative rule base with quality control behavior guidelines, a chain-like collaborative interaction strategy is generated to coordinate the quality control behaviors of all participating parties. This strategy specifies the concrete responsibilities, operational procedures, and collaboration methods of each participant in the quality control process, ensuring that all parties can conduct quality control according to the rules, forming an organic whole. Implementing this chain-like collaborative interaction strategy can improve the efficiency of quality control in meteorological observation projects and reduce the occurrence of quality problems.

[0066] Step 140: Convert the engineering quality monitoring data into a multi-dimensional dynamic view according to the chain-based collaborative interaction strategy, and perform abnormal pattern recognition on the quality indicator change trend view in the multi-dimensional dynamic view. Generate graded early warning information containing early warning level and early warning range based on the recognition results. The multi-dimensional dynamic view includes a quality control node association view, a quality impact transmission path view, and a quality indicator change trend view.

[0067] Based on the generated chain-like collaborative interaction strategy, the collected engineering quality monitoring data is transformed into multi-dimensional dynamic views. These views include a quality control node relationship view, a quality impact transmission path view, and a quality indicator change trend view. The quality control node relationship view displays the topological structure of relationships between key quality control nodes, helping users intuitively understand the relationships between nodes. The quality impact transmission path view uses animation to show the direction and intensity of quality impact transmission between different nodes, facilitating user analysis of the quality impact propagation process. The quality indicator change trend view displays the trend of quality indicators over time in the form of curve sequences, helping users promptly identify quality problems. Anomaly pattern recognition is performed on the quality indicator change trend view; by analyzing the indicator curve sequences, abnormal segments are identified. Based on the recognition results, tiered early warning information including warning levels and warning ranges is generated, providing timely warnings for quality control.

[0068] Step 141: Parse the permission collaboration rule base and quality control behavior guidance information in the chained collaborative interaction strategy, determine the multi-dimensional dynamic view access permissions and operation permissions corresponding to each participant, and generate view permission configuration parameters.

[0069] When converting engineering quality monitoring data into a multi-dimensional dynamic view, the first step is to parse the permission collaboration rule base and quality control behavior guidance information in the chained collaborative interaction strategy. Based on these rules and guidelines, the access permissions and operation permissions for each participant in the multi-dimensional dynamic view are determined. For example, some participants may only have viewing permissions for the view of the relationship between quality control nodes, while others may have access to and operation permissions for all views. This permission information is then organized and configured to generate view permission configuration parameters. These parameters can be used for data filtering and view generation, ensuring that each participant can only access and operate the views they are authorized to access.

[0070] Step 142: Based on the view permission configuration parameters and engineering quality monitoring data, perform data filtering and processing to extract the quality control node data, quality impact transmission path data, and quality indicator change data that each participant has the right to access, and generate a subset of view data.

[0071] Based on view permission configuration parameters, the engineering quality monitoring data undergoes data filtering. From the large amount of collected engineering quality monitoring data, data on quality control nodes, quality impact transmission paths, and changes in quality indicators that each participant has access to is extracted. For example, if a participant only has access to some key quality control nodes, only the data for those nodes is extracted. The extracted data is then organized and integrated to generate a subset of view data. This subset will serve as the basis for generating a multi-dimensional dynamic view, ensuring that each participant sees only the data they are authorized to access.

[0072] Step 143: According to the view type requirements of quality control node association, quality impact transmission path and quality indicator change trend, convert the view data subset into a multi-dimensional dynamic view containing node association topology, path flow animation and indicator curve sequence.

[0073] Based on a subset of view data, and according to view type requirements for quality control node relationships, quality impact transmission paths, and quality indicator change trends, the data is transformed into multi-dimensional dynamic views. For the quality control node relationship view, a topological graph structure with nodes as vertices and relationships as edges is generated based on the relationship parameters between nodes. This topological graph structure can intuitively display the relationships between key quality control nodes. For the quality impact transmission path view, the direction and speed of the animation flow arrows are set according to the path's direction and intensity parameters, generating a path visualization result containing dynamic flow arrows. For the quality indicator change trend view, the numerical changes of different quality indicators over time are converted into a continuous curve sequence. These curve sequences, with time as the horizontal axis and indicator values ​​as the vertical axis, clearly display the changing trends of quality indicators.

[0074] Step 1431: Perform topology construction processing on the quality control node data in the subset of view data, generate a topology graph structure with nodes as vertices and relationships as edges based on the association parameters between nodes, and determine the topology graph structure as the quality control node association relationship view.

[0075] When generating a multi-dimensional dynamic view, a topology structure is constructed for the quality control node data in the view data subset. The connections between nodes are determined based on the association parameters. A topology graph is constructed with nodes as vertices and association relationships as edges. For example, for different key quality control nodes in a meteorological observation project, such as equipment installation nodes, data acquisition nodes, and data processing nodes, the connection methods between nodes are determined based on their logical relationships and data flow. The constructed topology graph structure is then defined as the quality control node association relationship view. This view helps users intuitively understand the relationships between key quality control nodes, providing an important reference for quality control.

[0076] Step 1432: Perform flow direction animation generation processing on the quality influence transmission path data in the subset of view data. Set the direction and flow speed of the animation flow arrow according to the direction and intensity parameters of the path. Determine the path visualization result containing the dynamic flow arrow as the quality influence transmission path view.

[0077] For the quality impact propagation path data in the subset of view data, flow animation is generated. Based on the path's direction parameters, the direction of the animation's flow arrows is determined, indicating the direction of quality impact propagation. Based on the path's intensity parameters, the animation's flow speed is set; the greater the intensity, the faster the flow speed. In this way, the quality impact propagation path is displayed dynamically. The path visualization result containing the dynamic flow arrows is defined as the quality impact propagation path view. This view helps users intuitively observe the propagation process of quality impact between different nodes, facilitating the analysis of the root cause and scope of quality problems.

[0078] Step 1433: Perform time series transformation processing on the quality index change data in the subset of the view data, and convert the numerical changes of different quality indicators in the time dimension into a continuous curve sequence. The curve sequence contains a two-dimensional coordinate curve with time on the horizontal axis and index value on the vertical axis. The curve sequence is determined as a quality index change trend view.

[0079] The quality indicator change data in the subset of the view data undergoes time series transformation. The numerical changes of different quality indicators over time are converted into a continuous curve sequence. A two-dimensional coordinate curve is plotted with time on the horizontal axis and indicator values ​​on the vertical axis. For example, for the accuracy indicators of meteorological observation data, their values ​​at different time points are plotted on the curve, forming a continuous curve. These curve sequences are then used to create a quality indicator change trend view. This view helps users intuitively observe the changing trends of quality indicators over time and promptly identify signs of quality problems.

[0080] Step 1434: Based on the quality control node relationship view, quality impact transmission path view, and quality indicator change trend view, generate a multi-dimensional dynamic view with interactive switching function, which allows users to switch between different view types.

[0081] After generating separate views for quality control node relationships, quality impact transmission paths, and quality indicator change trends, these are integrated into a multi-dimensional dynamic view with interactive switching functionality. This interactive switching feature allows users to switch between different view types according to their needs. For example, a user can first view the quality control node relationship view to understand the relationships between nodes; then switch to the quality impact transmission path view to observe the transmission process of quality impacts; and finally switch to the quality indicator change trend view to analyze the changing trends of quality indicators. Through this interactive switching function, users can gain a more comprehensive understanding of the quality status of meteorological observation projects and improve the efficiency of quality control.

[0082] Step 144: Extract the indicator curve sequence of the quality indicator change trend view in the multi-dimensional dynamic view, perform pattern feature extraction processing on the indicator curve sequence, and obtain an indicator pattern feature set containing trend slope features, fluctuation frequency features, and abrupt change point distribution features.

[0083] To identify abnormal patterns in a quality indicator trend view, the sequence of indicator curves is first extracted. These curve sequences are then processed for pattern feature extraction. Through curve analysis, pattern features reflecting the curve's characteristics are extracted. These features include trend slope, fluctuation frequency, and abrupt change distribution. The trend slope indicates the degree of the curve's upward or downward trend, determined by calculating the curve's slope over different time periods. The fluctuation frequency indicates the curve's fluctuation pattern, determined by counting the number of fluctuations within a certain time period. The abrupt change distribution indicates the distribution of points of sudden change in the curve, determined by analyzing changes in the curve's derivative. These features are combined into an indicator pattern feature set for abnormal pattern identification.

[0084] Step 145: Perform matching and recognition processing between the indicator pattern feature set and the preset abnormal pattern feature library, identify abnormal segments in the indicator curve sequence that match the features in the abnormal pattern feature library, and generate abnormal pattern recognition results.

[0085] The extracted indicator pattern feature set is matched and identified against a pre-defined anomaly pattern feature library. This library contains various possible anomaly pattern features. By comparing the indicator pattern feature set with these features, abnormal segments in the indicator curve sequence that match features in the library are identified. For example, if the library defines an anomaly pattern of a sudden increase in trend slope, and a segment in the indicator curve sequence matches this feature, that segment is identified as an anomaly. The identified anomaly segments are recorded, and their corresponding anomaly pattern features are analyzed to generate anomaly pattern identification results. These results can serve as a reference for tiered early warning systems.

[0086] Step 1451: Perform feature evolution path extraction processing on each abnormal pattern feature template in the preset abnormal pattern feature library, extract the trend change sequence, fluctuation pattern sequence and mutation point occurrence order of each abnormal pattern in the time dimension, and generate a set of pattern feature evolution paths including trend evolution sub-path, fluctuation evolution sub-path and mutation evolution sub-path.

[0087] During the matching and recognition process, feature evolution paths are extracted from each anomaly pattern feature template in the pre-defined anomaly pattern feature library. Each anomaly pattern has a unique trend change sequence, fluctuation pattern sequence, and order of mutation points over time. By analyzing and extracting these features, a set of pattern feature evolution paths is generated, including trend evolution sub-paths, fluctuation evolution sub-paths, and mutation evolution sub-paths. The trend evolution sub-path represents the trend change of the anomaly pattern over time, the fluctuation evolution sub-path represents the fluctuation pattern change of the anomaly pattern, and the mutation evolution sub-path represents the order and temporal distribution of mutation points in the anomaly pattern. This set of pattern feature evolution paths can be used for path reasoning and comparison.

[0088] Step 1452: Perform time-series processing on the trend slope feature, fluctuation frequency feature and mutation point distribution feature in the indicator pattern feature set, and arrange each feature in the order of collection time to form a feature evolution sequence. The feature evolution sequence includes a trend change subsequence, a fluctuation pattern subsequence and a mutation point order subsequence.

[0089] The trend slope, fluctuation frequency, and abrupt change distribution features in the indicator pattern feature set are processed using time series analysis. These features are arranged chronologically according to their acquisition time, forming a feature evolution sequence. This sequence contains three subsequences: a trend change subsequence, a fluctuation pattern subsequence, and an abrupt change sequence subsequence. The trend change subsequence is a time-ordered sequence of trend slope features, the fluctuation pattern subsequence is a time-ordered sequence of fluctuation frequency features, and the abrupt change sequence subsequence is a time-ordered sequence of abrupt change distribution features. This time series analysis allows for better comparison and analysis with the pattern feature evolution path set.

[0090] Step 1453: Perform path reasoning on the trend change subsequence in the feature evolution sequence and the trend evolution subpath in the pattern feature evolution path set. By comparing the continuity and consistency of trend direction changes in the subsequence, the matching degree of the number of trend reversals, and the proportion of trend duration, identify abnormal candidate segments in the trend dimension.

[0091] The process involves path reasoning by comparing trend change subsequences in the feature evolution sequence with trend evolution subpaths in the pattern feature evolution path set. This includes comparing the continuity and consistency of trend direction changes within the subsequences (i.e., determining if the trend direction changes in the trend change subsequences are consistent with those in the trend evolution subpaths), comparing the matching degree of trend reversal numbers (i.e., determining if the number of trend reversals in the trend change subsequences is similar to that in the trend evolution subpaths), and comparing the proportion of different trend durations (i.e., determining if the proportion of different trend durations in the trend change subsequences is similar to that in the trend evolution subpaths). Through these comparisons and analyses, anomalous candidate segments in the trend dimension are identified. These anomalous candidate segments may represent parts of the indicator curve sequence that exhibit abnormal trend changes.

[0092] Step 1454: Perform morphological structure comparison processing between the wave morphology subsequence in the feature evolution sequence and the wave evolution subpath in the pattern feature evolution path set. By analyzing the morphological similarity of the wave cycle, the consistency of the variation law of the wave amplitude, and the correspondence of the wave peak occurrence position, identify abnormal candidate segments in the wave dimension.

[0093] The morphological structure of the fluctuation pattern subsequences in the feature evolution sequence is compared with that of the fluctuation evolution subpaths in the pattern feature evolution path set. The morphological similarity of the fluctuation cycle is analyzed, i.e., whether the fluctuation cycle shape in the fluctuation pattern subsequence is similar to that in the fluctuation evolution subpath. The consistency of the fluctuation amplitude variation pattern is analyzed, i.e., whether the fluctuation amplitude variation pattern in the fluctuation pattern subsequence is consistent with that in the fluctuation amplitude variation pattern in the fluctuation evolution subpath. The correspondence of the fluctuation peak positions is analyzed, i.e., whether the fluctuation peak positions in the fluctuation pattern subsequence correspond to the fluctuation peak positions in the fluctuation evolution subpath. Through these comparisons and analyses, abnormal candidate segments in the fluctuation dimension are identified. These abnormal candidate segments may be parts of the indicator curve sequence that exhibit abnormal fluctuations.

[0094] Step 1455: Perform sequence correlation analysis on the mutation point sequence subsequence in the feature evolution sequence and the mutation evolution subpath in the pattern feature evolution path set. By verifying the matching degree of the order of mutation points, the correlation of the time interval between adjacent mutation points, and the sequence consistency of mutation point types, identify abnormal candidate segments in the mutation dimension.

[0095] The sequence correlation analysis is performed between the mutation point sequence subsequence in the feature evolution sequence and the mutation evolution subpath in the pattern feature evolution path set. This involves verifying the matching degree of the mutation point occurrence order, i.e., determining whether the occurrence order of mutation points in the mutation point sequence subsequence is consistent with the occurrence order of mutation points in the mutation evolution subpath. It also verifies the correlation of time intervals between adjacent mutation points, i.e., determining whether the time intervals between adjacent mutation points in the mutation point sequence subsequence are similar to the time intervals between adjacent mutation points in the mutation evolution subpath. Finally, it verifies the sequence consistency of mutation point types, i.e., determining whether the sequence of mutation point types in the mutation point sequence subsequence is consistent with the sequence of mutation point types in the mutation evolution subpath. Through these verifications and analyses, anomalous candidate segments in the mutation dimension are identified. These anomalous candidate segments may be parts of the index curve sequence containing anomalous mutations.

[0096] Step 1456: Perform spatiotemporal intersection analysis on the anomalous candidate segments in the trend dimension, fluctuation dimension, and mutation dimension, and extract the anomalous candidate segments that appear in two or more dimensions at the same time as the first confidence anomalous segments, and the anomalous candidate segments that appear only in a single dimension as the second confidence anomalous segments.

[0097] After identifying anomalous candidate segments in the trend, fluctuation, and mutation dimensions, spatiotemporal intersection analysis is performed on these segments. Anomalous candidate segments appearing simultaneously in two or more dimensions are extracted and designated as first-confidence anomalous segments; these segments exhibit anomalousness across multiple dimensions and possess high confidence. Anomalous candidate segments appearing only in a single dimension are designated as second-confidence anomalous segments; these segments have relatively lower confidence. This spatiotemporal intersection analysis process enables more accurate identification of anomalous segments and improves the accuracy of anomalous pattern recognition.

[0098] Step 1457: Integrate the first confidence-based abnormal segments and the second confidence-based abnormal segments, record the start timestamp, end timestamp, involved abnormal dimensions, and corresponding abnormal pattern feature template identifiers of each abnormal segment, and generate an abnormal pattern recognition result containing the spatiotemporal distribution and pattern matching information of the abnormal segments.

[0099] The first and second confidence-level anomalous segments are integrated, and the start and end timestamps, involved anomalous dimensions, and corresponding anomalous pattern feature template identifiers of each anomalous segment are recorded. This information is then organized and summarized to generate an anomalous pattern recognition result that includes the spatiotemporal distribution and pattern matching information of anomalous segments. This result can comprehensively characterize the location, involved dimensions, and corresponding anomalous patterns of anomalous segments in the indicator curve sequence.

[0100] Step 146: Determine the warning level and warning range based on the abnormal segment location and abnormality degree parameters in the abnormal pattern recognition results, and generate graded warning information containing the warning level and warning range.

[0101] Based on the location and severity parameters of the abnormal segments in the anomaly pattern recognition results, the warning level and warning scope are determined. The location of the abnormal segment indicates the specific location where the anomaly occurs, and the severity parameter indicates the severity of the anomaly. Based on this information, the impact of the anomaly on project quality is assessed, thereby determining the warning level. Simultaneously, the warning scope is determined based on the corresponding nodes and downstream related nodes of the abnormal segment. The warning level and warning scope are integrated to generate a tiered warning information system that includes both the warning level and the warning scope. This information will provide timely warnings to quality control personnel so that appropriate measures can be taken to resolve quality issues.

[0102] Step 1461: Based on the spatiotemporal distribution and pattern matching information of the abnormal fragments in the abnormal pattern recognition results, extract the quality control node identifier corresponding to the abnormal fragment and the node level attribute of the node in the quality monitoring model of the whole process stage. The node level attribute is used to represent the core degree of the node in the quality control process.

[0103] When determining the warning level and scope, the first step is to extract the quality control node identifiers corresponding to the abnormal segments based on the spatiotemporal distribution and pattern matching information of the abnormal fragments in the anomaly pattern recognition results. These identifiers can accurately locate the nodes where the anomalies occur. Simultaneously, the node's hierarchical attribute within the full-process quality monitoring model is extracted. This attribute represents the node's core importance in the quality control process. For example, some nodes may be core nodes with a significant impact on project quality, while others may be non-core nodes with a relatively smaller impact. By extracting the node hierarchical attribute, the degree of impact of anomalies on project quality can be better assessed.

[0104] Step 1462: Trace the set of downstream associated nodes of the node corresponding to the abnormal segment through the inter-node quality influence transmission path unit, and analyze the propagation depth of the abnormal influence along the transmission path. The propagation depth is the maximum number of downstream nodes to which the abnormal influence can be transmitted.

[0105] Based on the identifier of the node corresponding to the anomalous fragment, the downstream associated node set of that node is traced through the quality impact propagation path unit between nodes. These downstream associated nodes may be affected by the quality of the anomalous node. The propagation depth of the anomalous impact along the propagation path is analyzed. Propagation depth refers to the maximum number of downstream nodes to which the anomalous impact can be propagated. For example, if the quality problem of an anomalous node can affect nodes two levels downstream, the propagation depth is 2. By analyzing the propagation depth, the scope and extent of the anomalous impact can be understood.

[0106] Step 1463: Count the number of nodes with core-level attributes in the downstream associated node set, and determine the coverage of the abnormal impact by combining the propagation depth. The coverage is positively correlated with both the number of core nodes and the propagation depth.

[0107] The number of nodes with core-level attributes in the downstream associated node set is counted, as these core nodes have a significant impact on project quality. The breadth of the anomaly's impact is determined by combining this with the propagation depth. The breadth of coverage is positively correlated with both the number of core nodes and the propagation depth; that is, the more core nodes and the greater the propagation depth, the wider the coverage. For example, if there are multiple core nodes in the downstream associated node set and the propagation depth is large, the breadth of the anomaly's impact will be relatively wide. By determining the breadth of coverage, the scope of the anomaly's impact on project quality can be assessed more comprehensively.

[0108] Step 1464: Construct an early warning level assessment logic based on the node hierarchy attributes, propagation depth, and coverage breadth. When the node corresponding to the abnormal segment is at the core level and the propagation depth exceeds the preset number of levels or the coverage breadth includes a preset number or more core nodes, it is determined to be at the first early warning level. When the node corresponding to the abnormal segment is at a non-core level and the propagation depth does not exceed the preset number of levels and the coverage breadth includes a preset number of core nodes, it is determined to be at the second early warning level. When the node corresponding to the abnormal segment is at a non-core level and the propagation depth is zero and the coverage breadth does not include any core nodes, it is determined to be at the third early warning level.

[0109] An early warning level assessment logic is constructed based on node hierarchy attributes, propagation depth, and coverage breadth. When the node corresponding to the abnormal segment is at the core level, and the propagation depth exceeds a preset number of levels, or the coverage breadth includes a preset number or more core nodes, it indicates a significant impact on project quality, and is classified as the first early warning level. When the node corresponding to the abnormal segment is at a non-core level, and the propagation depth does not exceed a preset number of levels, and the coverage breadth includes fewer than a preset number of core nodes, it indicates a relatively small impact on project quality, and is classified as the second early warning level. When the node corresponding to the abnormal segment is at a non-core level, the propagation depth is zero, and the coverage breadth does not include any core nodes, it indicates a very small impact on project quality, and is classified as the third early warning level. This early warning level assessment logic accurately determines the early warning level of an anomaly.

[0110] Step 1465: Based on the node corresponding to the abnormal segment and the set of downstream associated nodes, extract the associated path of the node corresponding to the abnormal segment in the quality impact transmission path unit between nodes, and determine the coverage of the set of downstream associated nodes and the associated path as the warning range. The warning range includes a node identifier list and a path identifier list.

[0111] Based on the set of nodes corresponding to the abnormal fragments and their downstream associated nodes, the associated paths of the nodes corresponding to the abnormal fragments in the quality impact transmission path units between nodes are extracted. These associated paths represent the transmission paths of the abnormal impact between nodes. The coverage area of ​​the downstream associated node set and associated paths is determined as the warning range, which includes a node identifier list and a path identifier list. The node identifier list lists the identifiers of the nodes affected by the abnormality, and the path identifier list lists the identifiers of the paths through which the abnormal impact is transmitted. By determining the warning range, the specific scope of the abnormal impact can be clearly defined, providing accurate information for quality control personnel to take appropriate measures.

[0112] Step 1466: Based on the warning level and the list of node identifiers and path identifiers in the warning range output by the warning level assessment logic, generate hierarchical warning information containing warning level field, core affected node field, associated path field and affected coverage level field.

[0113] Based on the warning level and the list of node and path identifiers within the warning scope output by the warning level assessment logic, hierarchical warning information is generated, including fields for warning level, core affected nodes, associated paths, and impact coverage level. The warning level field indicates the warning level of the anomaly; the core affected nodes field lists the identifiers of the core nodes affected by the anomaly; the associated paths field lists the identifiers of the paths through which the anomaly's impact propagates; and the impact coverage level field indicates the number of coverage levels of the anomaly's impact. By generating hierarchical warning information, important information about the anomaly can be clearly conveyed to quality control personnel, enabling timely measures to resolve quality issues.

[0114] As an optional embodiment, the method further includes:

[0115] Step 211: Perform real-time updates on the full-process stage quality monitoring model, adjust the quality attribute description information in the node quality feature description unit according to the newly collected engineering quality monitoring data, update the direction and intensity parameters of the directed graph structure in the inter-node quality influence transmission path unit, and synchronously correct the allowable fluctuation range in the quality control threshold interval unit.

[0116] As the meteorological observation project progresses, new engineering quality monitoring data will be continuously collected. To ensure the accuracy and effectiveness of the quality monitoring model throughout the entire process, it needs to be updated in real time. Based on the newly collected data, the quality attribute description information in the node quality characteristic description unit is adjusted. For example, if new data shows a change in the quality attribute of one of the key quality control nodes, the quality attribute description information of that node is updated accordingly. Simultaneously, the direction and intensity parameters of the directed graph structure in the quality influence transmission path unit between nodes are updated. If new data indicates a change in the quality influence transmission relationship between nodes, the direction and intensity of the directed graph structure are adjusted. Furthermore, the allowable fluctuation range in the quality control threshold interval unit is simultaneously corrected. Based on the statistical analysis of the new data and in conjunction with the engineering quality acceptance standards, the allowable fluctuation range of the quality attributes of each key quality control node is adjusted.

[0117] Step 212: Compare and analyze the differences between the updated full-process stage quality monitoring model and the original model, identify the changes in the node quality feature description unit, the quality influence transmission path unit between nodes, and the quality control threshold interval unit, and generate a model update difference report.

[0118] After updating the quality monitoring model across the entire process, a comparative analysis is performed between the updated model and the original model. The comparison is conducted on the node quality feature description units, the inter-node quality impact transmission path units, and the quality control threshold interval units to identify changes in these units. For example, in the node quality feature description units, the quality attribute description information of some nodes may have changed; in the inter-node quality impact transmission path units, the direction or strength of some directed connections may have changed; and in the quality control threshold interval units, the allowable fluctuation range of the quality attributes of some nodes may have been adjusted. These changes are compiled and recorded to generate a model update difference report, which will provide a reference for adjusting the chain-based collaborative interaction strategy.

[0119] Step 213: Adaptively adjust the chain-based collaborative interaction strategy according to the model update difference report. When the node quality feature description unit changes, re-process the mapping and matching of the node permission mapping relationship table; when the quality influence transmission path unit between nodes changes, re-process the association matching of the path permission association matrix; when the quality control threshold interval unit changes, update the conflict resolution triggering rules in the permission collaboration rule base.

[0120] Based on the model update difference report, adaptive adjustments are made to the chain-based collaborative interaction strategy. If the node quality feature description unit changes, the mapping and matching process of the node permission mapping table is re-performed. This is because changes in node quality features may affect the feature access permissions of each participant to the node, requiring a redefinition of these permissions. If the quality impact transmission path unit between nodes changes, the association matching process of the path permission association matrix is ​​re-performed. Changes in the path may affect the intervention permissions of each participant to the path, requiring a redefinition of these permissions. If the quality control threshold interval unit changes, the conflict resolution triggering rules in the permission collaboration rule base are updated. Adjustments to the quality control threshold may cause changes in permission conflict situations, requiring corresponding adjustments to the conflict resolution triggering rules.

[0121] Step 214: Send the adjusted chain-based collaborative interaction strategy to each participant in the matrix-based organizational collaborative architecture to ensure that the quality control behavior of each participant is consistent with the updated full-process stage quality monitoring model.

[0122] After adaptively adjusting the chain-based collaborative interaction strategy, the revised strategy is sent to all participants in the matrix-style organizational collaborative architecture. Each participant adjusts its quality control behavior according to the new chain-based collaborative interaction strategy. For example, participants may need to adjust feature access and operational permissions for key quality control nodes, or adjust the intervention methods for the quality impact propagation path. In this way, the quality control behavior of each participant is aligned with the updated end-to-end quality monitoring model, ensuring the effectiveness and accuracy of quality control.

[0123] Step 215: Based on the updated chain-based collaborative interaction strategy, the engineering quality monitoring data is reconstructed into a multi-dimensional dynamic view, and anomaly pattern recognition is performed on the quality indicator change trend view to generate updated hierarchical early warning information.

[0124] Based on the updated chain-based collaborative interaction strategy, the engineering quality monitoring data is reconstructed into a multi-dimensional dynamic view. Following the new permission configuration and operational specifications, the data is filtered and processed to generate a multi-dimensional dynamic view containing a view of the relationship between quality control nodes, a view of the transmission path of quality impacts, and a view of the changing trends of quality indicators. Anomaly pattern recognition is performed on the quality indicator changing trend view. Using the same method as before, pattern features of the indicator curve sequences are extracted and matched against a pre-set anomaly pattern feature library to identify anomalous segments. Based on the location and severity parameters of the anomalous segments, updated hierarchical early warning information is generated. This information reflects the latest engineering quality status and provides timely early warnings for quality control.

[0125] As an optional embodiment, the method further includes:

[0126] Step 311: Establish a quality impact tracing mechanism across quality control nodes. When an abnormal pattern appears in the quality indicator change trend view in the multi-dimensional dynamic view, trace the source node of the abnormal indicator based on the quality impact transmission path unit between nodes. The source node is a key quality control node located upstream of the abnormal indicator node in the quality impact transmission path.

[0127] To better address quality issues, a quality impact tracing mechanism across quality control nodes needs to be established. When anomalies appear in the trend view of quality indicators in a multi-dimensional dynamic view, the source node of the abnormal indicator's impact can be traced using the quality impact transmission path unit between nodes. The source node refers to the key quality control node upstream of the abnormal indicator node in the quality impact transmission path. By tracing the source node, the root cause of the quality problem can be identified, providing a basis for taking targeted measures.

[0128] Step 312: Perform in-depth analysis and processing on the engineering quality monitoring data of the traced-out impact source nodes, extract the quality attribute description information of the impact source nodes during the period of anomaly occurrence, analyze the correlation between this information and the quality attribute description information of the anomaly indicator nodes, and generate an impact correlation report.

[0129] After tracing the source node of the impact, in-depth analysis and processing of the engineering quality monitoring data for that node are performed. Quality attribute descriptions of the source node during the period of anomaly occurrence are extracted; this information reflects the quality status of the node at the time of the anomaly. The correlation between this information and the quality attribute descriptions of the anomaly indicator nodes is analyzed; for example, whether a change in one quality attribute of the source node leads to a corresponding change in the quality attribute of the anomaly indicator node. The analysis results are compiled and recorded to generate an impact correlation report, which details the quality correlation between the source node and the anomaly indicator node.

[0130] Step 313: Based on the impact correlation report and chain-based collaborative interaction strategy, coordinate the participants corresponding to the impact source node to carry out quality intervention processing, and send a quality intervention instruction containing the impact source node identifier, abnormal correlation information and intervention operation suggestions to the participant.

[0131] Based on impact correlation reporting and chain-based collaborative interaction strategies, the stakeholders corresponding to the impact source nodes are coordinated to conduct quality intervention. A quality intervention instruction is sent to the stakeholder, containing the impact source node identifier, anomaly correlation information, and intervention operation suggestions. The impact source node identifier is used to accurately locate the problem node, the anomaly correlation information explains the quality correlation between the impact source node and the anomalous indicator node, and the intervention operation suggestions guide the stakeholder to take specific measures to resolve the quality issue. By sending the quality intervention instruction, stakeholders are prompted to take timely action to resolve the quality problem.

[0132] Step 314: Receive the intervention result data fed back by the participants after they have performed the intervention operation according to the quality intervention instruction, input the intervention result data into the full-process stage quality monitoring model, and update the quality attribute description information affecting the source node in the node quality feature description unit.

[0133] After participating parties execute intervention operations according to quality intervention instructions, they will provide feedback on the intervention results data. This data is received and input into the end-to-end quality monitoring model. Based on the intervention results data, the quality attribute description information affecting the source node in the node quality characteristic description unit is updated. For example, if the intervention operation improves one of the quality attributes affecting the source node, the quality attribute description information of that node is updated accordingly. In this way, the end-to-end quality monitoring model can reflect the latest quality status of nodes in a timely manner.

[0134] Step 315: Track the trend of quality indicators after intervention for abnormal indicator nodes, monitor the recovery of indicators through the trend view of quality indicator changes in the multi-dimensional dynamic view, and generate intervention effect confirmation information when the indicators recover to the allowable fluctuation range in the quality control threshold interval unit, thus completing the tracing and intervention loop of cross-node quality impact.

[0135] The system tracks the quality indicator trends of abnormal nodes after intervention operations, and monitors the recovery of indicators through a multi-dimensional dynamic view of quality indicator trend changes. It observes whether the indicator curves gradually return to normal and whether they fall within the allowable fluctuation range of the quality control threshold interval. When the indicators recover to the allowable fluctuation range, it indicates that the intervention operation has been effective, generating intervention effect confirmation information. This information signifies that the tracing and intervention loop for cross-node quality impacts is complete, and the quality problem has been effectively resolved. Through this closed-loop management, the quality control level of meteorological observation projects can be continuously improved.

[0136] This application's embodiments start by collecting engineering quality monitoring data from key quality control nodes in meteorological observation projects, then constructing a full-process stage quality monitoring model, generating a chain-like collaborative interaction strategy, and finally realizing multi-dimensional dynamic view display of engineering quality monitoring data, anomaly pattern recognition, and hierarchical early warning. Overall, it solves the problems of fragmented quality monitoring and low collaborative efficiency caused by high technical barriers and poor process connection in meteorological observation engineering.

[0137] In detail, a full-chain, visualized quality control system built upon a chain-based dual-engine management approach digitizes key quality control points at each stage of the project, constructing a comprehensive quality monitoring indicator system. This transforms fragmented quality monitoring into a holistic system, comprehensively and systematically reflecting the project's quality status. Combined with a matrix-style organizational collaborative architecture, a cross-level, cross-departmental collaborative interaction mechanism, leveraging a chain-based collaborative interaction strategy, effectively improves the collaborative efficiency among all participants, breaks down departmental barriers, and ensures timely and accurate information transmission, guaranteeing collaborative work and jointly safeguarding project quality. A visualization model transforms quality data into multi-dimensional dynamic views, achieving a visualized presentation of quality status, process deviations, and rectification loops. This allows relevant personnel to intuitively understand the real-time status of project quality, promptly identify problems, and take corrective measures. Simultaneously, the designed quality early warning strategy intelligently identifies and classifies abnormal data based on preset thresholds, enabling timely detection of potential quality issues and tiered warnings according to the severity of the problems, improving the timeliness and effectiveness of quality control.

[0138] Based on the same inventive concept, embodiments of this application also provide a chain-based quality monitoring information visualization system. (See also...) Figure 2As shown, this is a schematic diagram of the structure of a possible chain-based quality monitoring information visualization system provided in an embodiment of this application. Figure 2 In the chain-type quality monitoring information visualization system 200, there are a processor 210 and a memory 220. The memory 220 stores computer programs that can be executed by the processor 210. By executing the instructions stored in the memory 220, the processor 210 can perform the steps of the chain-type quality monitoring information visualization method based on meteorological observation engineering described above.

[0139] Based on the same inventive concept, embodiments of this application provide a computer-readable storage medium including a computer program. When the computer program runs on a chain-based quality monitoring information visualization system, it causes the system to execute the steps of the aforementioned chain-based quality monitoring information visualization method based on meteorological observation engineering. In some possible implementations, various aspects of the chain-based quality monitoring information visualization method based on meteorological observation engineering provided in this application can also be implemented as a program product, including a computer program. When the program product runs on a chain-based quality monitoring information visualization system, it causes the system to execute the steps of the aforementioned chain-based quality monitoring information visualization method based on meteorological observation engineering. For example, the chain-based quality monitoring information visualization system can execute... Figure 1 The steps are shown in the figure.

[0140] The above content is only a preferred exemplary embodiment of this application and is not intended to limit the implementation of this application. Those skilled in the art can easily make corresponding modifications or alterations based on the main concept and spirit of this application.

Claims

1. A chain-based quality monitoring information visualization method based on meteorological observation engineering, characterized in that, The method includes: Collect engineering quality monitoring data generated at each key quality control node during the implementation of the meteorological observation project; Based on the engineering quality monitoring data, a full-process stage quality monitoring model reflecting the inherent correlation of each quality control node is constructed. This model includes node quality feature description units, quality impact transmission path units between nodes, and quality control threshold interval units. The process involves: identifying feature fields in the engineering quality monitoring data corresponding to each key quality control node and distinguishing data feature differences between different node types to generate a node data classification set; analyzing the key attributes reflecting quality status and their manifestations in the node data classification set to generate quality attribute description information for each key quality control node; organizing this information into structured feature vectors to generate node quality feature description units; performing temporal correlation analysis on the quality attribute description information of each key quality control node to capture the quality feature change sequence of different nodes over time; identifying potential correlation clues between nodes based on sequence comparison; and, based on these potential correlation clues and the logical dependencies of the engineering implementation process, analyzing the direction of quality impact between different key quality control nodes to construct initial directed connections between nodes; and further analyzing the data to generate node quality feature description units. Analyze the causal relationships of node feature changes in historical data, determine the influence degree parameters of each directed connection, and weight the initial directed connection relationships based on the influence degree parameters to construct a directed weighted graph structure. This directed weighted graph structure is then used as the quality influence transmission path unit between nodes. Historical quality compliance samples from the engineering quality monitoring data are screened, and statistical characteristic analysis is performed on the quality attribute data of each key quality control node in these historical quality compliance samples. The distribution pattern characteristics and dispersion index of the quality attribute data under different implementation stages are extracted. Based on the distribution pattern characteristics and the dispersion index, combined with the engineering quality acceptance standards, the allowable fluctuation range of the quality attributes of each key quality control node at different implementation stages is determined, and this allowable fluctuation range is used as the quality control threshold interval unit. After completing the multi-unit collaborative verification of the node quality feature description unit, the inter-node quality influence transmission path unit, and the quality control threshold interval unit, these units are integrated to generate a full-process stage quality monitoring model reflecting the inherent relationships among each quality control node. By combining the quality control permission information of each participant in the matrix-style organizational collaborative architecture, a correlation analysis is performed on the node quality feature description units and the quality impact transmission path units between nodes in the full-process stage quality monitoring model to generate a chain-like collaborative interaction strategy for coordinating the quality control behaviors of each participant: the quality control permission information of each participant in the matrix-style organizational collaborative architecture is divided into permission scopes to obtain a permission description set containing node control permission items and path intervention permission items; the node control permission items in the permission description set are mapped and matched with the node quality feature description units to determine the feature access rights of each participant to different key quality control nodes. The system generates a node permission mapping table; it then performs association matching between the path intervention permission items in the permission description set and the quality impact transmission path units between nodes to determine the intervention permission level of each participant for different quality impact transmission paths, generating a path permission association matrix; based on the node permission mapping table and the path permission association matrix, it performs permission conflict analysis to identify overlapping permission areas of different participants on the same node or path, generating permission conflict identification results; and based on the permission conflict identification results, it performs permission coordination processing on the node permission mapping table and the path permission association matrix to generate a chain-like collaborative interaction strategy for coordinating the quality control behaviors of each participant. The engineering quality monitoring data is converted into a multi-dimensional dynamic view according to the chain-based collaborative interaction strategy. Anomaly pattern recognition is performed on the quality indicator change trend view in the multi-dimensional dynamic view. Based on the recognition results, hierarchical early warning information including early warning level and early warning range is generated. The multi-dimensional dynamic view includes a quality control node association view, a quality impact transmission path view, and a quality indicator change trend view.

2. The method as described in claim 1, characterized in that, The permission conflict analysis based on the node permission mapping table and path permission association matrix identifies overlapping permission areas for different participants on the same node or path, generating permission conflict identification results, including: Traverse the node permission mapping table, extract multiple participants and their node control permission items corresponding to each key quality control node, compare the range intersection of the node control permission items of different participants, and mark the node permission conflict area when the range intersection is not empty. Traverse the path permission association matrix, extract multiple participants and their path intervention permission levels corresponding to each quality impact transmission path, compare the relationship between the path intervention permission levels of different participants, and mark the path permission conflict area when there is overlap of the same or higher level permissions. Collect all marked node permission conflict areas and path permission conflict areas, record the participant identifiers, conflict node or path identifiers, and conflict permission content involved in each conflict area, and generate permission conflict identification results containing conflict type, conflict location, and conflict participants.

3. The method as described in claim 1, characterized in that, The step involves performing permission coordination processing on the node permission mapping table and path permission association matrix based on the permission conflict identification results, generating a chain-like collaborative interaction strategy for coordinating the quality control behaviors of each participant, including: The node permission conflict areas in the permission conflict identification results are sorted by priority. The permission priority order is determined according to the hierarchical relationship of the participants in the matrix organizational collaborative architecture. The node control permission items of the first priority participants are retained as primary permissions, and the node control permission items of the second priority participants are adjusted to auxiliary permissions. The path permission conflict area in the permission conflict identification result is divided into intervention scope. According to the importance parameter of the quality impact transmission path, the path is divided into core path segment and non-core path segment. The intervention permission of the core path segment is assigned to the participant with the target path intervention permission level, and the intervention permission of the non-core path segment is assigned to other participants. A permission collaboration rule base is constructed based on the adjusted node permission mapping table and path permission association matrix. The permission collaboration rule base includes node permission allocation rules, path intervention collaboration rules, and conflict resolution triggering rules. Based on the permission collaboration rule base, quality control behavior guidance information for each participant is generated. The behavior guidance information includes node feature access process, path intervention operation specifications, and cross-participant collaborative response mechanism. By combining the aforementioned permission coordination rule base and quality control behavior guidance information, a chain-like collaborative interaction strategy is generated to coordinate the quality control behaviors of all participating parties.

4. The method as described in claim 1, characterized in that, The process involves converting the engineering quality monitoring data into a multi-dimensional dynamic view according to the chain-like collaborative interaction strategy, identifying abnormal patterns in the quality indicator change trend view of the multi-dimensional dynamic view, and generating graded early warning information including early warning level and early warning range based on the identification results, including: The permission collaboration rule base and quality control behavior guidance information in the chained collaborative interaction strategy are analyzed to determine the multi-dimensional dynamic view access permissions and operation permissions of each participant, and view permission configuration parameters are generated. Based on the view permission configuration parameters and engineering quality monitoring data, data filtering and processing are performed to extract the quality control node data, quality impact transmission path data and quality indicator change data that each participant has the right to access, and generate a subset of view data. According to the view type requirements of quality control node association, quality impact transmission path and quality indicator change trend, the view data subset is converted into a multi-dimensional dynamic view containing node association topology, path flow animation and indicator curve sequence. Extract the indicator curve sequence of the quality indicator change trend view in the multi-dimensional dynamic view, and perform pattern feature extraction processing on the indicator curve sequence to obtain an indicator pattern feature set containing trend slope features, fluctuation frequency features, and abrupt change point distribution features. The indicator pattern feature set is matched and identified with a preset abnormal pattern feature library to identify abnormal segments in the indicator curve sequence that match the features in the abnormal pattern feature library, and an abnormal pattern identification result is generated. Based on the location and degree of abnormality parameters of the abnormal segments in the abnormal pattern recognition results, the warning level and warning range are determined, and hierarchical warning information containing the warning level and warning range is generated.

5. The method as described in claim 4, characterized in that, According to the view type requirements of quality control node association, quality impact transmission path, and quality indicator change trend, the subset of view data is converted into a multi-dimensional dynamic view containing node association topology, path flow animation, and indicator curve sequence, including: The quality control node data in the subset of view data is processed to construct a topology structure. Based on the association parameters between nodes, a topology graph structure with nodes as vertices and association relationships as edges is generated. The topology graph structure is then determined as the quality control node association relationship view. The quality impact transmission path data in the subset of view data is processed to generate flow direction animation. The direction and speed of the flow arrow in the animation are set according to the direction and intensity parameters of the path. The path visualization result containing the dynamic flow arrow is determined as the quality impact transmission path view. The quality indicator change data in the subset of the view data is processed by time series transformation, and the numerical changes of different quality indicators in the time dimension are converted into a continuous curve sequence. The curve sequence contains a two-dimensional coordinate curve with time on the horizontal axis and indicator value on the vertical axis. The curve sequence is determined as a quality indicator change trend view. Based on the quality control node relationship view, quality impact transmission path view, and quality indicator change trend view, a multi-dimensional dynamic view with interactive switching function is generated, which allows users to switch between different view types.

6. The method as described in claim 4, characterized in that, The step of matching and identifying the indicator pattern feature set with a preset abnormal pattern feature library, identifying abnormal segments in the indicator curve sequence that match features in the abnormal pattern feature library, and generating abnormal pattern identification results includes: The feature evolution path extraction process is performed on each feature template of the pre-set abnormal pattern feature library. The trend change sequence, fluctuation pattern sequence and mutation point occurrence order of each abnormal pattern in the time dimension are extracted to generate a set of pattern feature evolution paths containing trend evolution sub-path, fluctuation evolution sub-path and mutation evolution sub-path. The trend slope feature, fluctuation frequency feature, and mutation point distribution feature in the indicator pattern feature set are processed by time series processing. Each feature is arranged in the order of collection time to form a feature evolution sequence. The feature evolution sequence includes a trend change subsequence, a fluctuation pattern subsequence, and a mutation point order subsequence. The trend change subsequence in the feature evolution sequence is compared with the trend evolution subpath in the pattern feature evolution path set for path reasoning. By comparing the continuity and consistency of trend direction changes in the subsequence, the matching degree of the number of trend turning points, and the proportion of trend duration, abnormal candidate segments in the trend dimension are identified. The wave morphology subsequence in the feature evolution sequence is compared with the wave evolution subpath in the pattern feature evolution path set. By analyzing the morphological similarity of the wave cycle, the consistency of the variation law of the wave amplitude, and the correspondence of the wave peak occurrence position, abnormal candidate segments in the wave dimension are identified. The sequence of mutation points in the feature evolution sequence is subjected to sequence correlation analysis with the mutation evolution sub-paths in the pattern feature evolution path set. By verifying the matching degree of the order of mutation points, the correlation of the time interval between adjacent mutation points, and the sequence consistency of mutation point types, abnormal candidate segments in the mutation dimension are identified. Spatiotemporal intersection analysis was performed on the anomalous candidate segments in the trend dimension, fluctuation dimension, and mutation dimension. Anomalous candidate segments that appeared in two or more dimensions simultaneously were extracted as anomalous segments with first confidence, and anomalous candidate segments that appeared only in a single dimension were extracted as anomalous segments with second confidence. Integrate the first confidence-level anomalous segments and the second confidence-level anomalous segments, record the start timestamp, end timestamp, involved anomalous dimensions, and corresponding anomalous pattern feature template identifiers for each anomalous segment, and generate an anomalous pattern recognition result containing the spatiotemporal distribution and pattern matching information of the anomalous segments.

7. The method as described in claim 4, characterized in that, The step of determining the warning level and warning range based on the abnormal segment location and abnormality degree parameters in the abnormal pattern recognition result, and generating hierarchical warning information containing the warning level and warning range, includes: Based on the spatiotemporal distribution and pattern matching information of the abnormal fragments in the abnormal pattern recognition results, the quality control node identifier corresponding to the abnormal fragment and the node hierarchy attribute of the node in the quality monitoring model of the whole process are extracted. The node hierarchy attribute is used to represent the core degree of the node in the quality control process. By tracing the set of downstream associated nodes of the node corresponding to the abnormal segment through the quality influence transmission path unit between nodes, the propagation depth of the abnormal influence along the transmission path is analyzed. The propagation depth is the maximum number of downstream nodes that the abnormal influence can be transmitted to. The number of nodes with core-level attributes in the downstream associated node set is counted, and the coverage of the abnormal impact is determined by combining the propagation depth. The coverage is positively correlated with both the number of core nodes and the propagation depth. The warning level assessment logic is constructed based on the node hierarchy attributes, propagation depth and coverage breadth. When the node corresponding to the abnormal segment is at the core level and the propagation depth exceeds the preset number of levels or the coverage breadth includes more than the preset number of core nodes, it is determined to be the first warning level. When the node corresponding to the abnormal segment is a non-core level, the propagation depth does not exceed the preset number of levels, and the coverage breadth includes fewer core nodes than the preset number, it is determined to be the second warning level. When the node corresponding to the abnormal fragment is a non-core level, the propagation depth is zero, and the coverage breadth does not include core nodes, it is determined to be the third warning level; Based on the set of nodes corresponding to the abnormal fragments and their downstream associated nodes, the associated paths of the nodes corresponding to the abnormal fragments in the inter-node quality impact transmission path unit are extracted, and the coverage of the set of downstream associated nodes and associated paths is determined as the warning range. The warning range includes a list of node identifiers and a list of path identifiers. Based on the warning level and the list of node identifiers and path identifiers in the warning range output by the warning level assessment logic, hierarchical warning information containing warning level field, core affected node field, associated path field and affected coverage level field is generated.

8. A chain-based quality monitoring information visualization system, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of any one of the methods described in claims 1 to 7.

Images