Landscape planning progress tracking method and system

Abstract

The invention relates to the technical field of data reasoning, and discloses a landscape planning progress tracking method and system, and the method comprises the steps: packaging a timestamp and a geographic coordinate tag on a multi-modal original data set in a landscape planning process, and obtaining a space-time tagging original data package; eliminating noise and inconsistency in the space-time tagging original data packet; constructing a project progress state graph; identifying an attribute difference value between the project progress state map and a pre-stored project plan digital benchmark; mapping the attribute difference value to a historical multi-layer progress deviation identification rule base, judging a root source and an influence level, and generating a structured deviation analysis report; self-adaptive early warning and decision support information is generated by matching case features in the structured deviation analysis report and the historical coping strategy knowledge base; self-adaptive early warning and decision support information is transmitted to a landscape planning terminal, and a project plan digital reference is driven to be visually updated; the landscape planning progress tracking efficiency can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data reasoning, in particular to a landscape planning progress tracking method and system. BACKGROUND

[0002] The prior art has significant deficiencies in the data collection and processing link of landscape planning progress tracking. The multi-modal raw data is not standardized and integrated, and the spatial and temporal labels are not encapsulated. Only single type data or data without format uniformity is scattered collected, resulting in lack of spatio-temporal correlation and consistency of the data, and the data set cannot form a complete reflection of the project progress. At the same time, the project progress state atlas is not constructed by repairing data missing and eliminating logical contradictions. The progress is only presented by simple data list or table statistics, which cannot intuitively reflect the spatio-temporal and logical correlation between entities, and cannot accurately capture the overall situation of the project progress, so the quality of the basic data provided for subsequent deviation identification is low.

[0003] The prior art has significant defects in the deviation analysis and decision support link of landscape planning progress. The systematic deviation identification and analysis mechanism is not established, and the deviation is only judged by manual comparison of the planned and actual progress. The root cause and influence level of the deviation cannot be accurately located, and the analysis results lack structure and scientificity. The adaptive decision support information is not generated combined with the historical response strategy knowledge base, and the adjustment scheme is only formulated depending on experience. It is difficult to adapt to the dynamic changes of the project requirements, and the warning information and the plan benchmark are not updated synchronously, resulting in delayed decision response and untimely plan adjustment. The project progress cannot be effectively controlled, and the landscape planning project cannot be promoted as expected. SUMMARY

[0004] The present application provides a landscape planning progress tracking method and system to solve the problems raised in the background art.

[0005] To achieve the above purpose, the present application provides a landscape planning progress tracking method, comprising: S1, collecting a multi-modal raw data set in the landscape planning process, and encapsulating time stamp and geographic coordinate label in the multi-modal raw data set to obtain a spatio-temporal labeled raw data package of the multi-modal raw data set; S2, eliminating noise and inconsistency in the spatio-temporal labeled raw data package, and taking entities in the spatio-temporal labeled raw data package as nodes and taking spatio-temporal and logical correlation between entities as edges to construct a project progress state atlas of the spatio-temporal labeled raw data package; S3, identifying attribute difference values of the project progress state atlas and the pre-stored project plan digital benchmark; S4, mapping the attribute difference value to a historical multi-layer progress deviation identification rule library to determine the root source and impact level of the deviation and generate a structured deviation analysis report of the project progress status graph; S5, generating adaptive early warning and decision support information for the landscape planning process by matching the structured deviation analysis report with case features in a historical coping strategy knowledge base; S6, transmitting the adaptive early warning and decision support information to a landscape planning terminal and synchronously driving the pre-stored project plan digital benchmark to perform visual update.

[0006] In a preferred embodiment, a multi-modal raw data set in the landscape planning process is collected, and a time and space tagged raw data package of the multi-modal raw data set is obtained by encapsulating a time stamp and a geographic coordinate label on the multi-modal raw data set, comprising: integrating image data, text description data, and numerical measurement data of the landscape planning area into a multi-modal raw data set of the landscape planning area; eliminating data format differences of the multi-modal raw data set to obtain a standardized multi-modal data set of the landscape planning area; adding a time stamp and a geographic coordinate label to a data item in the multi-modal raw data set to obtain a time and space tagged raw data package of the multi-modal raw data set.

[0007] In a preferred embodiment, the noise and inconsistency in the time and space tagged raw data package are eliminated, and an project progress status graph of the time and space tagged raw data package is constructed by taking entities in the time and space tagged raw data package as nodes and taking time and space and logical association relationships between entities as edges, comprising: repairing data items with missing data, inconsistent formats, and logical contradictions in the time and space tagged raw data package to obtain a standardized data package of the time and space tagged raw data package; performing tensor synthesis on time dimension features and space dimension features in the standardized data package to obtain a time and space feature vector of the standardized data package; calculating a time and space association matrix between entities in the standardized data package according to the time and space feature vector; analyzing task dependency relationships between the entities to establish a logical association matrix of the standardized data package; performing multi-dimensional feature fusion on the time and space association matrix and the logical association matrix to obtain a comprehensive association matrix of the standardized data package; constructing an project progress status graph of the time and space tagged raw data package by taking the entities as nodes and taking element values of the comprehensive association matrix as edge weights.

[0008] In a preferred embodiment, the formula for calculating the spatio-temporal correlation strength in the spatio-temporal correlation matrix is: ; In the formula, is the spatio-temporal correlation strength of entity and entity , is the standard deviation of Gaussian distribution, is a natural constant, is a time dimension weight factor, is a space dimension weight factor, is the timestamp difference between entity and entity , is the Euclidean distance between the geographic coordinates of entity and entity , is a time scale parameter, is a space scale parameter.

[0009] In a preferred embodiment, the identification of the attribute difference value between the project progress status graph and the pre-stored project plan digital benchmark includes: extracting the current state attribute set of the entity in the project progress status graph; retrieving the planned state attribute set of the entity in the pre-stored project plan digital benchmark; comparing the current state attribute set with the planned state attribute set on an entity-by-entity basis to identify a list of state attribute inconsistent entities in the landscape planning process; determining the attribute difference value between the current state attribute and the planned state attribute based on the entities in the list of state attribute inconsistent entities.

[0010] In a preferred embodiment, the mapping of the attribute difference value to the historical multi-layer progress deviation identification rule library to determine the root cause and impact level of the deviation and generating a structured deviation analysis report for the project progress status graph includes: analyzing the entity identifier and deviation type information in the attribute difference value; performing rule matching between the attribute difference value and the root cause determination rule in the historical multi-layer progress deviation identification rule library to identify the root cause of the deviation; based on the identified root cause, querying the impact level classification rule in the historical multi-layer progress deviation identification rule library to determine the impact level of the deviation; integrating the root cause and the impact level into a structured deviation analysis report for the project progress status graph.

[0011] In a preferred embodiment, the generating adaptive early warning and decision support information for the landscape planning process by matching the structured deviation analysis report with case features in a historical response strategy knowledge base comprises: performing deep semantic analysis on the structured deviation analysis report to obtain a multi-dimensional feature vector of the structured deviation analysis report; in the historical response strategy knowledge base, matching historical cases with the highest situational fit degree with the multi-dimensional feature vector to generate a high-quality case candidate set for the landscape planning process; evaluating the response timeliness, effect persistence and resource adaptability of cases in the high-quality case candidate set to screen out preferred cases in the high-quality case candidate set that meet the requirements; based on the dynamic context environment of the current landscape planning project, performing performance fine-tuning on the response strategy in the preferred case to obtain an enhanced response strategy for the preferred case; integrating the enhanced response strategy and real-time risk assessment results to generate adaptive early warning and decision support information for the landscape planning process.

[0012] In a preferred embodiment, the performance fine-tuning of the response strategy in the preferred case based on the dynamic context environment of the current landscape planning project to obtain an enhanced response strategy for the preferred case comprises: analyzing the dynamic context environment of the current landscape planning project and extracting key environmental parameters of the current landscape planning project; establishing a matching degree evaluation system for the key environmental parameters and the response strategy in the preferred case, and identifying adaptation differences between strategy elements and environmental requirements in the matching degree evaluation system; based on the adaptation differences, performing element-level optimization and adjustment on the response strategy to obtain environmental adaptation factors of the response strategy; verifying the expected execution effect of the environmental adaptation factors in the dynamic context environment to screen out effective adaptation factors that meet the effectiveness threshold; integrating the effective adaptation factors into the response strategy to obtain an enhanced response strategy for the preferred case.

[0013] In a preferred embodiment, the transmitting the adaptive early warning and decision support information to the landscape planning terminal and synchronously driving the pre-stored project plan digital benchmark to perform visual update comprises: formatting the adaptive early warning and decision support information according to a preset information format to obtain a standardized early warning and decision data package of the adaptive early warning and decision support information; transmitting the standardized early warning and decision data package to the landscape planning terminal; The landscape planning terminal analyzes the standardized early warning decision data packet and extracts early warning content and decision suggestion information in the standardized early warning decision data packet; According to the early warning content and decision suggestion information, the pre-stored project plan digital benchmark is driven to be visually refreshed.

[0014] In order to solve the above problems, the present application also provides a landscape planning progress tracking system, the system comprises: A data acquisition label module is used to acquire a multi-modal original data set in a landscape planning process, and encapsulate a time stamp and a geographic coordinate label on the multi-modal original data set to obtain a spatio-temporal labeled original data packet of the multi-modal original data set. A data cleaning graph module is used to eliminate noise and inconsistency in the spatio-temporal labeled original data packet, and construct a project progress state graph of the spatio-temporal labeled original data packet by taking entities in the spatio-temporal labeled original data packet as nodes and taking spatio-temporal and logical association relationships between entities as edges. A difference identification calculation module is used to identify attribute difference values of the project progress state graph and a pre-stored project plan digital benchmark. A deviation analysis report module is used to map the attribute difference values to a historical multi-layer progress deviation identification rule library, determine the root source and influence level of the deviation, and generate a structured deviation analysis report of the project progress state graph. A decision support generation module is used to generate adaptive early warning and decision support information of the landscape planning process by matching the structured deviation analysis report and case features in a historical coping strategy knowledge base. An information synchronization update module is used to deliver the adaptive early warning and decision support information to a landscape planning terminal and synchronously drive the pre-stored project plan digital benchmark to be visually updated.

[0015] Compared with the prior art, the present application has the following beneficial effects: 1. The present application provides accurate state presentation for landscape planning progress tracking through multi-modal data integration and progress graph construction. Image, text and numerical measurement data in the landscape planning process are collected, time stamp and geographic coordinate labels are encapsulated after standardization processing to form a spatio-temporal labeled data packet, data missing and logical contradictions are repaired, spatio-temporal association and logical association are fused to construct a project progress state graph, and the association relationship and current progress state between entities are clearly presented to provide comprehensive and reliable basic data support for progress deviation identification.

[0016] 2.The application significantly improves the efficiency and control ability of landscape planning progress tracking by means of precise deviation analysis and adaptive decision support. By comparing the progress state atlas with the planned digital benchmark for each entity, the attribute difference value is accurately identified. The deviation source and influence level are determined by mapping the historical rule library, and a structured analysis report is generated. The adaptive early warning and decision support information are generated by matching the historical response strategies and optimizing adjustment combined with the dynamic context environment. The information is pushed to the terminal and drives the plan benchmark visualization update, realizing the rapid response and precise control of progress deviation, and ensuring the project to proceed according to the plan. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A flowchart of a landscape planning progress tracking method provided by an embodiment of the application is shown in the figure. Figure 2 A functional module diagram of a landscape planning progress tracking system provided by an embodiment of the application is shown in the figure. The implementation of the object of the application, functional characteristics and advantages will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0018] It should be understood that the specific embodiments described herein are only used to explain the application and not to limit the application.

[0019] An embodiment of the application provides a landscape planning progress tracking method. The execution subject of the landscape planning progress tracking method includes but is not limited to at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the application. In other words, the landscape planning progress tracking method can be executed by software or hardware installed in a terminal device or a server device. The server includes but is not limited to a single server, a server cluster, a cloud server, or a cloud server cluster, etc. The server can be a stand-alone server, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content distribution networks (CDN), and big data and artificial intelligence platforms, etc. basic cloud computing services.

[0020] Referring to Figure 1 A flowchart of a landscape planning progress tracking method provided by an embodiment of the application is shown in the figure. In this embodiment, the landscape planning progress tracking method includes: S1, collecting a multi-modal original data set in the landscape planning process, and encapsulating a time stamp and a geographic coordinate label on the multi-modal original data set to obtain a spatio-temporal labeled original data packet of the multi-modal original data set; In the embodiment of the present application, the multi-modal original data set in the landscape planning process is collected, and a time stamp and a geographic coordinate label are packaged for the multi-modal original data set to obtain a spatio-temporal labeled original data package of the multi-modal original data set, comprising: integrating image data, text description data and numerical measurement data of the landscape planning area into a multi-modal original data set of the landscape planning area; eliminating data format differences of the multi-modal original data set to obtain a standardized multi-modal data set of the landscape planning area; adding a time stamp and a geographic coordinate label to the data items in the multi-modal original data set to obtain a spatio-temporal labeled original data package of the multi-modal original data set.

[0021] When integrating image data, text description data and numerical measurement data of the landscape planning area into a multi-modal original data set of the landscape planning area, different types of basic data in the landscape planning area are comprehensively collected, image data covers visual information such as topography, existing vegetation, building distribution, water body shape, etc. in the region, and through high-definition shooting equipment, different time periods and different angles are collected to ensure that the overall situation of the planning area is covered. The text description data includes the historical background, functional positioning, planning target, ecological protection requirements and other related explanatory contents of the region, which are systematically extracted from the pre-planning research report and regional development planning documents.

[0022] The numerical measurement data includes terrain elevation data, soil physical and chemical property data, illumination intensity data, water resource reserve data and other quantifiable indicators, which are obtained through professional measurement equipment on-site survey. The three types of data are classified and arranged according to data sources and collection scenarios, and repeated and invalid data are removed to ensure that each type of data completely covers the core information required for landscape planning, and finally integrated to form a multi-modal original data set containing vision, text and quantitative indicators.

[0023] When eliminating the data format differences of the multi-modal original data set to obtain a standardized multi-modal data set of the landscape planning area, the format types of each type of data in the multi-modal original data set are sorted out one by one, the image data may have different pixel specifications and file formats, the text description data may have different encoding formats and document types, and the numerical measurement data may have different units and precision standards.

[0024] For image data, a unified pixel size and file format are used for conversion to ensure that the resolution and color mode of all images remain consistent. For text description data, texts in different encoding formats are uniformly converted to standard encoding, and different types of documents are converted to a unified text format, while removing format redundant information in the text.

[0025] For numerical measurement data, the values in different units are converted to standard units, and the values with different precisions are calibrated to ensure that the data precision meets the planning and analysis requirements. After format standardization processing, all types of data maintain uniformity in format, forming a standardized multi-modal data set with a structured specification that can be directly used for subsequent processing.

[0026] When adding time stamps and geographic coordinate labels to data items in the multi-modal raw data set, the collection time and collection location information of each data item are determined. The time stamp is accurately labeled according to the year, month, day, hour, minute, and second of data actual collection. For continuously collected numerical measurement data, the time stamps are labeled one by one according to the collection interval. For image data and text description data, the specific time point of collection or generation is labeled.

[0027] The geographic coordinate label is based on a global unified geographic coordinate system and is labeled based on the latitude, longitude, and altitude information recorded by the collection device, ensuring that each data item can be accurately corresponded to a specific geographic location within the landscape planning area. For image data covering a certain range, the geographic coordinate boundaries of the shooting range are labeled. For specific areas involved in text description data, the geographic coordinate information of the area is labeled. The labeled time stamps and geographic coordinate labels are bound to each corresponding data item to ensure that the data items correspond one-to-one with the space-time information, ultimately forming a space-time labeled raw data package containing the data itself and the corresponding space-time attributes.

[0028] The beneficial effects are that the image, text description, and numerical measurement data of the landscape planning area are integrated, and the three types of data provide visual intuitive information, textual description information, and quantitative index information, fully covering the basic data dimensions required for landscape planning, avoiding the limitations of single-type data, and providing rich and complete raw data support for subsequent progress tracking.

[0029] The format differences in the multi-modal raw data set are eliminated, and the image, text, and numerical data are uniformly formatted to ensure consistency in specifications, encoding, units, and other aspects, solving the compatibility problem of data from different sources and ensuring that the data can be directly used for subsequent integration and analysis, improving data processing efficiency.

[0030] Adding time stamps and geographic coordinate labels to data items enables each data item to have clear time and space attributes, establishing a space-time association between data and the landscape planning process, allowing data to accurately correspond to specific planning stages and regional locations, and providing key attribute support for subsequent construction of project progress status atlases and identification of space-time association relationships.

[0031] The finally formed spatio-temporal tagged original data packet has a standardized format and contains complete spatio-temporal attributes, has clear data structure and logical coherence, and can be directly used for subsequent operations such as noise elimination and correlation matrix construction, greatly reducing the processing difficulty of subsequent progress tracking links and laying a foundation for improving the efficiency of landscape planning progress tracking.

[0032] S2, eliminating noise and inconsistency in the spatio-temporal tagged original data packet, and constructing a project progress state atlas of the spatio-temporal tagged original data packet with entities in the spatio-temporal tagged original data packet as nodes and spatio-temporal and logical correlation between entities as edges; In the embodiment of the application, the eliminating noise and inconsistency in the spatio-temporal tagged original data packet, and constructing a project progress state atlas of the spatio-temporal tagged original data packet with entities in the spatio-temporal tagged original data packet as nodes and spatio-temporal and logical correlation between entities as edges, comprises: repairing data items with missing data, inconsistent format and logical contradiction in the spatio-temporal tagged original data packet to obtain a standardized data packet of the spatio-temporal tagged original data packet; tensor synthesizing time dimension features and space dimension features in the standardized data packet to obtain a spatio-temporal feature vector of the standardized data packet; calculating a spatio-temporal correlation matrix between entities in the standardized data packet according to the spatio-temporal feature vector; analyzing task dependency relationship between the entities to establish a logical correlation matrix of the standardized data packet; multi-dimensional feature fusion of the spatio-temporal correlation matrix and the logical correlation matrix to obtain a comprehensive correlation matrix of the standardized data packet; constructing a project progress state atlas of the spatio-temporal tagged original data packet with the entities as nodes and element values of the comprehensive correlation matrix as edge weights.

[0033] The calculation formula of the spatio-temporal correlation strength in the spatio-temporal correlation matrix is: ; In the formula, is the spatio-temporal correlation strength of entity and entity , is the standard deviation of Gaussian distribution, is a natural constant, is a time dimension weight factor, is a space dimension weight factor, is the timestamp difference between entity and entity , is the timestamp difference between entity and entity Euclidean distance between geographical coordinates, is a time scale parameter, is a space scale parameter.

[0034] In repairing the data missing, inconsistent format and logical contradiction of the data items in the spatio-temporal tagged original data package, the standardization data package of the spatio-temporal tagged original data package is obtained. Each data item in the data package is checked one by one to check whether there is information blank data missing. For the missing data items, the distribution rule of the same type of data and the characteristics of the associated data are reasonably supplemented to ensure that the supplemented data is logically consistent with the overall data. Check whether the format of all data items meets the standardization requirements, and uniformly convert the formats that do not meet the requirements to keep the data uniform in file type, encoding method, numerical unit, etc.

[0035] The logical relationship between the data items is sorted out, and the conflicting data is identified, such as time sequence contradiction, geographical coordinates and description position inconsistency, etc. The contradictory data is corrected by checking the original collection records, referring to the associated data, etc. Finally, the standardization data package with complete data, uniform format and logically self-consistent is formed.

[0036] In tensor synthesis of the time dimension characteristics and the space dimension characteristics in the standardization data package, the time dimension characteristics of each entity in the standardization data package are extracted, including data collection time, task plan start and end time, progress node time, etc. related to time information, and are arranged in time sequence to form a time characteristic sequence. The space dimension characteristics of each entity are extracted, including geographical coordinates, regional attribution, spatial distribution relationship, etc. related to location information, and are arranged in spatial correlation logic to form a spatial characteristic sequence. The time characteristic sequence and the spatial characteristic sequence are deeply fused in the form of tensor synthesis, and the core correlation between time and space characteristics is retained in the fusion process, so that the spatio-temporal information of each entity forms a unified feature set, and finally the set is converted into a spatio-temporal feature vector which can accurately represent the spatio-temporal attributes of the entity.

[0037] According to the spatio-temporal feature vector, the spatio-temporal correlation matrix between entities in the standardization data package is calculated. The spatio-temporal feature vectors corresponding to the entities in the standardization data package are selected one by one, and the time information and space information in the spatio-temporal feature vectors of any two entities are compared. The correlation degree in the time dimension is analyzed, including the interval of the data collection time of the two entities, the overlapping range of the task execution time, etc. The correlation degree in the space dimension is analyzed, including the distance between the geographical coordinates of the two entities, the overlapping area of the spatial distribution, etc. The correlation degrees are quantified by a fixed evaluation method, and the quantification results are filled into the corresponding positions of the matrix according to the corresponding relationship of the entities. Each element in the matrix represents the spatio-temporal correlation strength of the corresponding two entities, and finally a complete spatio-temporal correlation matrix is formed.

[0038] The task dependency relationship between entities is analyzed, the logical association matrix of the standardized data package is established, the specific tasks corresponding to each entity in the standardized data package are sorted out, and the execution content, target requirement and precondition of each task are clarified. By analyzing the preconditions and subsequent impacts of the tasks, the dependency relationship between entities is determined, such as a task corresponding to an entity must be executed after the completion of a task corresponding to another entity, or the tasks corresponding to two entities need to be promoted and cooperated synchronously. The tightness of these dependency relationships is quantified, such as strong dependency, weak dependency, no dependency, etc. The quantification results are organized into a matrix according to the corresponding relationship of the entities, and each element in the matrix represents the logical association strength between the corresponding two entities, thereby establishing a logical association matrix.

[0039] In order to obtain the comprehensive association matrix of the standardized data package by multi-dimensional feature fusion of the space-time association matrix and the logical association matrix, the elements in each matrix that can reflect the core information of entity association are retained based on the two matrices. For the element value at each matrix position, the importance in space-time association and logical association is weighted and integrated, so that the fused element value can not only reflect the space-time association strength between entities, but also reflect the logical dependency degree. In the fusion process, the structure of the matrix is ensured to remain consistent, and the corresponding relationship of the entities does not change. In this way, the two single-dimensional association matrices are integrated into a comprehensive association matrix that can comprehensively reflect the comprehensive association between entities.

[0040] In order to construct the project progress state graph of the space-time labeled original data package, each entity in the standardized data package is taken as an independent node in the graph, and a unique identifier is assigned to each node to distinguish different entities.

[0041] According to the element value in the comprehensive association matrix, the weight of the edge between nodes is determined. The larger the element value, the higher the comprehensive association strength between the corresponding two entities, and the greater the weight of the edge. When the element value is zero, it means that there is no association between the two entities, and no edge is established between the nodes. According to the identifier of the node and the weight of the edge, the connection relationship between the nodes in the graph is established, and it is ensured that the weight of each edge accurately corresponds to the corresponding element value in the comprehensive association matrix, and finally the project progress state graph is formed, which can intuitively present the association relationship and association strength between entities.

[0042] The standard deviation of the Gaussian distribution is determined according to the overall distribution characteristics of the space-time association strength between entities in the historical data. By statistically analyzing a large number of association strength values of entity pairs, an index reflecting the dispersion degree of these values is calculated, which is used as the specific value of the parameter.

[0043] Natural constant is a fixed constant in mathematics, whose value is fixed and unchangeable, used in the formula to construct the exponential function form to achieve the nonlinear mapping of the correlation strength.

[0044] The time dimension weight factor is set according to the importance of the time factor in the landscape planning project to the influence of the entity correlation. If the time difference has a greater influence on the entity correlation, a larger value is given, and vice versa. The specific size is determined through expert evaluation combined with the actual needs of the project.

[0045] The space dimension weight factor is set according to the importance of the space factor in the landscape planning project to the influence of the entity correlation. If the spatial distance has a more significant influence on the entity correlation, a larger value is set, otherwise a smaller value is set. The specific size is determined through expert evaluation combined with the actual situation of the project.

[0046] Entity Entity The timestamp difference between entity and entity is the timestamp information of the two entities directly extracted from the spatiotemporal tagged original data packet. The difference between the two timestamps is calculated, which reflects the interval between the two entities in time.

[0047] Entity Entity The Euclidean distance between the geographic coordinates of entity and entity is obtained from the spatiotemporal tagged original data packet. Based on the calculation method of Euclidean distance, the straight-line distance between the two coordinate points in space is calculated to obtain the distance value, which reflects the interval between the two entities in space.

[0048] The time scale parameter is determined according to the time period of the landscape planning project and the time rhythm of entity activity, which is used to measure the proportion of the time difference relative to the time scale of the project. The specific value is set by analyzing the overall time span and key time nodes of the project.

[0049] The space scale parameter is determined according to the spatial range of the landscape planning area and the spatial characteristics of entity distribution, which is used to measure the proportion of the spatial distance relative to the space scale of the project. The specific value is set by analyzing the geographical range of the planning area and the spatial distribution density of entities.

[0050] The meaning of the formula is to consider the time difference and spatial distance between entities, and calculate an index that can quantitatively reflect the correlation strength between the two entities in the time-space dimension, i.e. the spatiotemporal correlation strength between entity and entity .

[0051] In the calculation process, first, the ratio of the timestamp difference value and the time scale parameter is multiplied by the time dimension weight factor, the ratio of the Euclidean distance between the geographic coordinates and the space scale parameter is multiplied by the space dimension weight factor, then the square of the sum of the two results is multiplied by negative one-half, the result is taken as the index of the exponential function, the exponential function value is calculated based on the natural constant, and finally the product of the standard deviation of the Gaussian distribution and the square root of pi is divided by the product, and the result is the spatiotemporal association strength of the entity and the entity The greater the strength value, the closer the association of the two entities in space and time, and vice versa.

[0052] The beneficial effect is that by repairing data missing, inconsistent formats and logical contradictions in the spatiotemporal tagged original data packet, solving the problem of incomplete and non-standard data, the generated standardized data packet can accurately reflect the actual data situation of landscape planning, providing reliable and unified data basis for subsequent association analysis and graph construction, avoiding progress tracking deviation caused by data problems.

[0053] The time and space dimension characteristics of the standardized data packet are tensor synthesized to fuse the dispersed spatiotemporal information into a unified spatiotemporal feature vector, completely retaining the association relationship of entities in time evolution and spatial distribution, making the spatiotemporal properties of entities more intuitive and easy to analyze, and providing clear feature basis for calculating the spatiotemporal association strength between entities.

[0054] The spatiotemporal association matrix between entities is calculated based on the spatiotemporal feature vector, the matrix elements accurately quantify the spatiotemporal association strength of any two entities, clearly presenting the close degree of the association of entities in time interval and spatial distance, and providing quantifiable spatiotemporal association data support for subsequent integration of logical association and construction of graph.

[0055] The task dependency relationship between entities is analyzed and a logical association matrix is established, the logical relationship such as task execution order, pre-post association and the like corresponding to the entities is converted into quantifiable matrix elements, the mutual influence of entities in task promotion is clearly presented, the core logical association information other than spatiotemporal association is supplemented, and the entity association analysis is more comprehensive.

[0056] The spatiotemporal association matrix and the logical association matrix are fused with multi-dimensional features, the spatiotemporal contact and task logical dependency of entities are comprehensively considered, and the generated comprehensive association matrix can comprehensively reflect the overall association between entities, avoid one-sidedness of single-dimensional association analysis, and provide scientific and comprehensive basis for graph edge weight setting.

[0057] By constructing a project progress status graph with entities as nodes and comprehensive association matrix elements as edge weights, the abstract entity association data is transformed into an intuitive graph structure. This allows for a clear view of the current status of each entity and clarifies the association strength and dependency relationships between entities, enabling staff to quickly grasp the overall project progress and providing an intuitive visualization tool for subsequent deviation identification and progress control.

[0058] The beneficial effects are that the formula integrates two core spatiotemporal parameters, timestamp difference and Euclidean distance of geographic coordinates, and combines them with time and space dimension weighting factors. This not only reflects the basic relationship between time intervals and spatial distances between entities, but also highlights the differences in the influence of different dimensions on the strength of the relationship through weighting factors. This adapts to the differentiated role of spatiotemporal factors on entity relationships in landscape planning, making the quantitative results more in line with actual scenarios.

[0059] By introducing time scale parameters and spatial scale parameters, timestamp differences and Euclidean distances are scaled to avoid deviations in the calculation of correlation strength caused by differences in the overall time period and spatial scope of the project. For example, in large-scale landscape planning projects, the scale parameters can be adjusted to adapt to long-term and wide-ranging spatiotemporal correlation assessments, ensuring that the calculation standards for correlation strength of projects of different scales are consistent, and the results are more comparable and reasonable.

[0060] Based on the Gaussian distribution, a computational model is constructed. The probability density function is built using the natural constant and standard deviation, so that the spatiotemporal correlation strength presents a nonlinear distribution that conforms to the laws of reality. The smaller the spatiotemporal difference between entities, the higher the correlation strength. The greater the difference, the stronger the correlation weakens reasonably. This accurately simulates the objective laws of spatiotemporal correlation between entities in landscape planning, avoids the distortion of correlation strength caused by linear calculation, and makes the quantitative results more scientific.

[0061] The formula can directly output the specific values ​​of the spatiotemporal correlation strength between entities. These values ​​can be directly used as elements of the spatiotemporal correlation matrix, providing accurate quantitative data support for subsequent fusion with the logical correlation matrix and construction of the project progress status map. This ensures that the matrix can truly reflect the spatiotemporal correlation of entities and lays a reliable data foundation for setting the edge weights of the map and subsequent progress analysis.

[0062] S3. Identify the attribute differences between the project progress status map and the pre-stored project plan digital benchmark; In this embodiment of the invention, identifying the attribute difference values ​​between the project progress status map and the pre-stored project plan digital benchmark includes: Extract the current state attribute set of entities in the project progress status map; Retrieve the set of planned status attributes for entities in the pre-stored project plan digital baseline; The current state attribute set and the planned state attribute set are compared entity by entity to identify a list of state attribute inconsistent entities of the landscape planning process; According to the entities in the list of state attribute inconsistent entities, attribute difference values between the current state attribute and the planned state attribute are determined.

[0063] When extracting the current state attribute set of entities in the project progress state graph, all entities in the project progress state graph are traversed, and the core state information of each entity is collected one by one, including the task completion progress, the current execution phase, the resource input situation, the space-time coordinate update state and other attributes directly related to the project progress. The attribute information of each entity collected is classified and arranged to ensure that each attribute can accurately reflect the current actual state of the entity. The attribute information of all entities is classified and integrated according to the entity identifier to form a current state attribute set that completely covers the current state of all entities in the graph.

[0064] When retrieving the planned state attribute set of entities in the pre-stored project plan digital benchmark, it is clear that the pre-stored project plan digital benchmark is a standard data set prepared in the early stage of the landscape planning project that contains the planned requirements of each entity. According to the unique identifier of the entity in the project progress state graph, the planned information corresponding to each entity is accurately retrieved in the pre-stored project plan digital benchmark.

[0065] The planned state attribute of each entity is extracted, including the planned completion progress, the preset execution phase, the planned resource input, the planning space-time coordinate, and other attributes. These planned attribute information is ordered and arranged according to the entity identifier to ensure consistency with the entity ordering and attribute categories of the current state attribute set, and finally a planned state attribute set is formed.

[0066] When comparing the current state attribute set and the planned state attribute set entity by entity to identify a list of state attribute inconsistent entities of the landscape planning process, the entity identifier is used as the basis for association, and each attribute of each entity in the current state attribute set is compared with the same attribute of the corresponding entity in the planned state attribute set one by one. Whether the current attribute value and the planned attribute value are completely consistent is checked one by one. If there is any attribute value that does not match, such as the actual completion progress does not meet the planned requirements, the current execution phase does not match the preset phase, etc., the entity is included in the marked list. After completing the attribute-by-attribute comparison of all entities, the marked entities are summarized to form a list of state attribute inconsistent entities that clearly lists all entities whose state attributes do not match the plan.

[0067] According to the entity in the state attribute inconsistent entity list, when determining the attribute difference value between the current state attribute and the planned state attribute, the inconsistent attribute item in the current state attribute set of each entity in the state attribute inconsistent entity list and the corresponding planned attribute item in the planned state attribute set are called again. For each pair of inconsistent attribute items, the difference is quantified, such as the difference between the actual completion ratio and the planned completion ratio for progress type attributes, the difference between the actual input and the planned input for resource input type attributes, and the deviation between the actual coordinates and the planned coordinates for space-time type attributes. The quantification results of each inconsistent attribute item of each entity are integrated to form an attribute difference value that can comprehensively reflect the difference between the current and planned state of the entity, ensuring that each inconsistent entity has a corresponding accurate attribute difference value.

[0068] The beneficial effects are that when extracting the current state attribute set of the entity in the project progress state graph, the core attributes such as task completion progress, resource input situation, and space-time coordinate update state of the entity are comprehensively collected, ensuring that the attribute set can completely and truly reflect the current actual state of the entity, providing accurate "actual value" reference for subsequent comparison with planned attributes, and avoiding deviation in difference identification caused by missing or one-sided current state information.

[0069] When retrieving the planned state attribute set of the entity in the pre-stored project plan digital benchmark, the corresponding planned information is accurately matched according to the unique identifier of the entity, and the planned attributes such as planned completion progress, preset execution stage, and planned space-time coordinates are extracted to form a clear "planned value" standard, ensuring that the comparison between the current and planned attributes is conducted under the same dimension and standard, avoiding comparison failure caused by inconsistent standards.

[0070] Comparing the current state attribute set with the planned state attribute set entity by entity and attribute by attribute can accurately identify any entity with inconsistent attributes, forming a state attribute inconsistent entity list, avoiding missing entities with deviations, and focusing the range of subsequent difference quantification on inconsistent entities, reducing invalid calculations, and improving difference identification efficiency.

[0071] For the entities in the inconsistent entity list, the difference between the current value and the planned value of the inconsistent attribute item is analyzed, and the difference is quantified according to the attribute type to obtain an attribute difference value that can accurately reflect the specific situation of each entity deviating from the plan, providing accurate quantitative data support for subsequent positioning of the root cause of the deviation and evaluation of the impact level, and avoiding inaccurate decision-making caused by general deviation analysis.

[0072] S4, mapping the attribute difference value to a historical multi-layer progress deviation identification rule library to determine the root cause and impact level of the deviation, and generating a structured deviation analysis report of the project progress state graph; In the embodiment of the present application, the attribute difference value is mapped to the historical multi-layer progress deviation identification rule library, the root cause and impact level of the deviation are determined, and a structured deviation analysis report of the project progress status graph is generated, which includes: The entity identifier and deviation type information in the attribute difference value are analyzed; The attribute difference value is matched with the root cause determination rule in the historical multi-layer progress deviation identification rule library to identify the root cause of the deviation; Based on the identified root cause, the impact level classification rule in the historical multi-layer progress deviation identification rule library is queried to determine the impact level of the deviation; The root cause and the impact level are integrated into the structured deviation analysis report of the project progress status graph.

[0073] When analyzing the entity identifier and deviation type information in the attribute difference value, the core information corresponding to each attribute difference value is disassembled one by one. The entity identifier is extracted from the entity unique identifier associated with the attribute difference value to ensure accurate correspondence with the specific entity in the project progress status graph. The deviation type is defined by analyzing the formation cause and manifestation form of the attribute difference value, such as progress deviation caused by insufficient resource input, position deviation caused by time and space coordinate deviation, quality deviation caused by non-compliance with execution standards, etc. Different types of deviation characteristics are clearly distinguished, and the extracted entity identifier and defined deviation type are arranged according to the corresponding relationship to provide a clear information basis for subsequent rule matching.

[0074] When matching the attribute difference value with the root cause determination rule in the historical multi-layer progress deviation identification rule library to identify the root cause of the deviation, it is clear that the historical multi-layer progress deviation identification rule library stores corresponding rules of various deviation types and root causes. These rules are formed based on a large number of historical project deviation cases.

[0075] Taking the disassembled deviation type as an index, the corresponding root cause determination rule is retrieved in the rule library, and the consistency of the specific characteristics of the attribute difference value and the rule description is compared. For example, when the deviation type is insufficient resource input, the rule description about resource shortage and unreasonable allocation of the root cause in the rule library is matched. When the attribute difference value characteristics completely match a certain rule, the source corresponding to the rule is determined as the root cause of the deviation, ensuring that the identification of the root cause conforms to the historical experience rules and rule requirements.

[0076] When determining the impact level of the deviation based on the identified root cause, the historical multi-layer progress deviation identification rule library has classified clear impact level classification rules according to the impact degree of the root cause on project progress, quality, cost, etc., including slight impact, general impact, and serious impact.

[0077] According to the identified deviation root cause, locate the corresponding impact level classification rule in the rule base, analyze the chain reaction, impact range and the degree of hindering the overall goal of the project that the root cause may cause, such as the shortage of core resources, which is judged as a serious impact level according to the rule; the deviation caused by local execution details is judged as a slight impact level, and it is ensured that the judgment of the impact level is highly consistent with the rule description.

[0078] When integrating the root cause and the impact level into the structured deviation analysis report of the project progress status map, first determine the fixed framework of the structured report, including the core modules of entity identification, deviation type, root cause, impact level and associated suggestions. The entity identification corresponding to each attribute difference value, the decomposed deviation type, the identified root cause and the determined impact level are filled in the corresponding modules of the report one by one, ensuring that each information item is accurately corresponding and without omission. The report content is logically combed, sorted by entity importance or deviation impact level, so that the report structure is clear and the levels are distinct, facilitating users to quickly obtain key deviation information, and finally forming a structured deviation analysis report that can comprehensively and systematically present the project progress deviation situation.

[0079] The beneficial effects are that when analyzing the entity identification and deviation type information in the attribute difference value, the unique identification of the corresponding entity and the specific performance type of the deviation are extracted from the attribute difference value, such as whether the deviation is caused by insufficient resource input or the deviation caused by deviation of time and space coordinates. This process can accurately lock the entity object with deviation and the core characteristics of the deviation, providing a clear analysis starting point for subsequent matching of root cause judgment rules and determination of impact level, avoiding confusion in deviation analysis direction caused by ambiguous information.

[0080] When matching the attribute difference value with the root cause judgment rule in the historical multi-layer progress deviation identification rule base, relying on the corresponding rules in the rule base based on a large number of historical project deviation cases, such as the rules of resource shortage, unreasonable allocation and other root causes corresponding to resource insufficient deviation, by comparing the consistency of attribute difference value characteristics and rule description, the root cause of the deviation can be accurately identified, avoiding subjective deviation caused by relying only on artificial experience to determine the root cause, making the deviation tracing more objective and accurate.

[0081] Based on the identified root cause, query the impact level classification rule in the rule base, and combine the potential impact of the root cause on the project progress, quality, cost and other aspects, such as the root cause of core resource shortage corresponding to the serious impact level, and the root cause of local execution detail deviation corresponding to the slight impact level, which can scientifically determine the impact level of the deviation and clearly define the degree of harm that the deviation may cause, providing risk level basis for subsequent development of coping strategies, avoiding too general or inaccurate risk assessment.

[0082] When integrating the root cause and impact level into a structured deviation analysis report, the identification, deviation type, root cause and impact level of each deviation entity are sorted according to a fixed framework to form a report with clear logic and complete content, so that workers can quickly obtain key information of the deviation and accurately grasp the overall situation of the deviation. Compared with scattered deviation information, the structured report can more efficiently provide comprehensive data support for subsequent matching of historical coping strategies and generation of decision support information, thereby improving the decision response efficiency.

[0083] S5, generating adaptive early warning and decision support information of the landscape planning process by matching the structured deviation analysis report with case features in a historical coping strategy knowledge base; In the embodiment of the present application, the adaptive early warning and decision support information of the landscape planning process is generated by matching the structured deviation analysis report with case features in a historical coping strategy knowledge base, which includes: performing deep semantic analysis on the structured deviation analysis report to obtain a multi-dimensional feature vector of the structured deviation analysis report; In the historical coping strategy knowledge base, matching historical cases with the highest situational fit degree with the multi-dimensional feature vector to generate a high-quality case candidate set of the landscape planning process; Evaluating the response timeliness, effect persistence and resource adaptability of the cases in the high-quality case candidate set to select preferred cases in the high-quality case candidate set that meet the requirements; Based on the dynamic context environment of the current landscape planning project, the performance of the coping strategies in the preferred cases is fine-tuned to obtain enhanced coping strategies of the preferred cases; Integrating the enhanced coping strategies with real-time risk assessment results to generate adaptive early warning and decision support information of the landscape planning process.

[0084] The performance of the coping strategies in the preferred cases is fine-tuned based on the dynamic context environment of the current landscape planning project to obtain enhanced coping strategies of the preferred cases, which includes: Analyzing the dynamic context environment of the current landscape planning project to extract key environmental parameters of the current landscape planning project; Establishing a matching degree evaluation system of the key environmental parameters and the coping strategies in the preferred cases, and identifying the adaptation differences between the strategy elements and environmental requirements in the matching degree evaluation system; Based on the adaptation differences, the coping strategies are optimized and adjusted at the element level to obtain environmental adaptation factors of the coping strategies; Verifying the expected execution effect of the environmental adaptation factors in the dynamic context environment to select effective adaptation factors that meet the efficiency threshold; Integrate the effective adaptation factor into the coping strategy to obtain an enhanced coping strategy of the preferred case.

[0085] When performing deep semantic analysis on the structured bias analysis report to obtain a multi-dimensional feature vector of the structured bias analysis report, the core semantic information in the report is extracted sentence by sentence, including key contents such as the specific task corresponding to the entity identifier, the core features of the bias type, the key causes of the root source, the quantitative performance of the influence level, etc. The extracted semantic information is classified and sorted according to dimensions such as bias attributes, influence ranges, and cause features, and the information of each dimension is converted into a representable feature item to ensure that each feature item accurately reflects the core semantics of the report. All dimension feature items are integrated in a fixed order to form a multi-dimensional feature vector that can comprehensively cover the key information of the report, providing accurate feature basis for subsequent case matching.

[0086] In the historical coping strategy knowledge base, the historical cases with the highest situational fit degree are matched with the multi-dimensional feature vector to generate a high-quality case candidate set for the landscape planning process. The historical bias coping cases of a large number of landscape planning projects are stored in the historical coping strategy knowledge base, and each case contains a corresponding feature vector and a coping strategy.

[0087] With the current obtained multi-dimensional feature vector as a reference, the feature vectors of each historical case in the knowledge base are compared one by one to analyze the fit degree of the two in key dimensions such as bias type, root source, influence level, and involved entities. A number of historical cases with the highest fit degree are selected, these cases have a high similarity in bias situation to the current project bias situation, and the coping strategies have high reference value, these cases are summarized to form a high-quality case candidate set.

[0088] The coping timeliness, effect sustainability, and resource adaptability of the cases in the high-quality case candidate set are evaluated, and the preferred cases that meet the requirements in the high-quality case candidate set are selected. For each candidate case, the time span from the implementation of the coping strategy to the generation of the effect is analyzed to determine whether it can quickly solve the bias problem of the current project, thereby evaluating the coping timeliness.

[0089] The effect sustainability is evaluated by investigating whether the bias is controlled for a long time and whether the project can continue to advance according to the plan after the implementation of the coping strategy in the case. The resource adaptability is evaluated by comparing the resources required by the coping strategy in the case with the resource reserves of the current project, such as manpower, material resources, and financial resources, to determine whether the resources are matched. A unified evaluation standard is set, only the cases with strong coping timeliness, good effect sustainability, and high resource adaptability are retained, and finally the preferred cases that meet the requirements are selected.

[0090] When the performance of the coping strategy in the preferred case is fine-tuned based on the dynamic context of the current landscape planning project, the dynamic context of the current project includes real-time information such as the remaining resources of the current project, the progress pace, the priority of unfinished tasks, and external environmental changes.

[0091] The implementation conditions, operation steps, and resource requirements of the coping strategy in the preferred case are analyzed one by one. Combined with the dynamic context of the current project, the parts of the strategy that do not match the current environment are adjusted, such as optimizing the resource allocation ratio to adapt to the current resource reserve, adjusting the time node of the implementation steps to fit the current progress pace, and supplementing the adaptation measures to cope with external environmental changes. After targeted fine-tuning, the coping strategy is more in line with the actual situation of the current project, the performance is improved, and an enhanced coping strategy is formed.

[0092] When integrating the enhanced coping strategy and real-time risk assessment results to generate adaptive early warning and decision support information for the landscape planning process, real-time risk assessment results are based on current project deviation and external environmental changes to identify potential derivative risks, risk diffusion range, and impact.

[0093] Fusion of the specific implementation steps, resource allocation scheme, and expected effect of the enhanced coping strategy with the risk points, warning thresholds, and prevention suggestions in the real-time risk assessment results can help avoid derivative risks while implementing the coping strategy. The information is integrated according to the structure of "deviation coping measures + risk warning prompts + decision execution suggestions" to ensure clear logic and focus. Ultimately, adaptive early warning and decision support information is formed to provide real-time guidance for landscape planning projects.

[0094] When analyzing the dynamic context of the current landscape planning project and extracting key environmental parameters of the current landscape planning project, the real-time state of the current project is comprehensively sorted out, including the current progress of the project, the total amount of resources invested and the remaining resource reserve, the priority of unfinished tasks, external policy requirements and industry standard specifications, real-time changes in natural environmental conditions, and other environmental information related to the implementation of the coping strategy. These environmental information is classified and screened, and secondary information unrelated to the adaptation of the coping strategy is removed. The core content that directly affects the execution effect of the strategy is retained, and these core content is converted into key environmental parameters that can be clearly described and used for adaptation evaluation, ensuring that each parameter accurately reflects the core environmental characteristics of the current project.

[0095] When establishing the matching degree evaluation system of key environmental parameters and coping strategies in the preferred case, and identifying the adaptation differences between the strategy elements and environmental requirements in the matching degree evaluation system, the core elements of the coping strategies in the preferred case are first disassembled, including the type and quantity of resources required for strategy implementation, the time node requirements of execution steps, applicable task scenario conditions, and policy and standard bottom lines that need to be met.

[0096] Based on the key environmental parameters, a matching degree evaluation system is constructed, including resource adaptation, time adaptation, scenario adaptation, and standard adaptation dimensions, each of which corresponds to a specific evaluation direction. Each element of the coping strategy is compared with the corresponding environmental requirements in the evaluation system, and whether the element meets the specific requirements of the current environmental parameters is analyzed, such as whether the resources required by the strategy exceed the remaining resources of the current project, whether the execution time node conflicts with the current progress, etc. The adaptation differences between each element and the environmental requirements that do not fit are accurately identified.

[0097] Based on the adaptation differences, the elements of the coping strategy are optimized and adjusted to obtain the environmental adaptation factors of the coping strategy. For each identified adaptation difference, a corresponding element optimization scheme is developed. If there is a resource adaptation difference, the use ratio of resources in the strategy is adjusted, the resource type is replaced, or the resource allocation method is optimized according to the current project's resource reserve; if there is a time adaptation difference, the execution steps of the strategy are adjusted in terms of sequence or time node in combination with the current project progress rhythm; if there is a scenario or standard adaptation difference, the implementation details of the strategy are modified to meet the scenario characteristics and the latest standard requirements of the current project. Each optimization adjustment content is extracted separately as an environmental adaptation factor that can independently act on the coping strategy. Each adaptation factor corresponds to a specific adaptation difference solution, ensuring that the factor can accurately fill the adaptation gap between the strategy and the environment.

[0098] When verifying the expected execution effect of the environmental adaptation factor in the dynamic context environment and screening out the effective adaptation factors that meet the efficiency threshold, the dynamic context environment of the current project is simulated, each environmental adaptation factor is introduced into the corresponding coping strategy element, and the possible results after the implementation of the factor are deduced. The deduced results are evaluated to determine whether they can solve the corresponding adaptation differences, whether they will cause new adaptation problems, and whether they can improve the overall execution effect of the coping strategy, such as whether the resource class adaptation factor can match the current reserve with the resource demand of the strategy, and whether the time class adaptation factor can make the strategy execution fit the current progress. Clear efficiency judgment standards are set. Only the adaptation factors that achieve the preset effect requirements and effectively improve the strategy adaptation are recognized as meeting the efficiency threshold, and all effective adaptation factors that meet the requirements are finally screened out.

[0099] When integrating the effective adaptation factors into the coping strategies to obtain the enhanced coping strategies of the preferred cases, each effective adaptation factor is correspondingly integrated into the relevant elements of the strategy according to the original structure and logic of the coping strategy. It is ensured that the adaptation factor seamlessly connects with the original content of the strategy and does not damage the overall integrity and execution logic of the strategy, such as integrating the resource adaptation factor into the resource configuration module of the strategy and embedding the time adaptation factor into the execution step planning of the strategy. After the integration is completed, the strategy is combed as a whole to check the synergy between the elements, so as to ensure that the optimized strategy can be smoothly executed in the dynamic context environment and the adaptability and execution efficiency are significantly improved, and finally the enhanced coping strategy of the preferred case is formed.

[0100] The beneficial effect is that when the structured deviation analysis report is deeply semantically analyzed, the core information such as entity identification, deviation type, root source, impact level and the like in the report is comprehensively disassembled, and these information is converted into a multi-dimensional feature vector according to the deviation attribute, impact range, cause characteristics and the like. The vector completely retains the key semantics of the report and can accurately represent the core situation of the current deviation, providing standardized and comparable feature basis for subsequent matching of similar cases in the historical coping strategy knowledge base, avoiding deviation caused by incomplete feature extraction.

[0101] When matching the historical cases with the highest situational fit degree of the multi-dimensional feature vector in the historical coping strategy knowledge base, the multi-dimensional features of the current deviation are taken as the benchmark, and the feature vectors of the historical cases in the knowledge base are compared one by one, focusing on the consistency of the key dimensions such as deviation type, root source and impact level. The high-quality case candidate set selected has a high similarity in deviation situation with the current project, and the corresponding coping strategy has been verified by historical practice and has high reference value, providing a reliable case basis for subsequent selection of preferred cases and generation of effective coping strategies.

[0102] When evaluating the coping timeliness, effect persistence and resource adaptability of the cases in the high-quality case candidate set, the execution effect time, long-term effect stability and resource demand of the coping strategy in the case are analyzed respectively. Only the cases that meet the standards in the three dimensions are selected as the preferred cases, so as to avoid selecting cases with slow response, short-term effect or resource demand exceeding the current project capacity, and to ensure that the coping strategy of the preferred case can adapt to the basic needs of the current project and provide a high-quality template for subsequent strategy fine-tuning.

[0103] When fine-tuning the coping strategies of the preferred cases based on the dynamic context of the current landscape planning project, real-time information such as the remaining resources, progress rhythm, and external environmental changes of the current project is combined to adjust elements in the strategy that do not match the current environment, such as optimizing resource allocation ratios, adjusting execution time nodes, and supplementing environmental adaptation measures. The enhanced coping strategies that have been fine-tuned can accurately fit the actual scenario of the current project, avoiding the problem of insufficient adaptability caused by directly applying historical strategies, and significantly improving the execution efficiency of the strategies.

[0104] When integrating the enhanced coping strategies with real-time risk assessment results, the specific implementation steps and resource allocation schemes of the strategies are combined with the derived risks and prevention suggestions identified in real time to form adaptive early warning and decision support information that includes "deviation coping measures + risk warning prompts + decision execution suggestions". This information not only clearly outlines the solution path for deviations, but also provides early warning of potential risks, providing comprehensive decision guidance for staff and achieving dynamic and adaptive control of landscape planning progress, effectively ensuring that the project progresses as planned.

[0105] The beneficial effects are that when analyzing the dynamic context of the current landscape planning project, real-time information such as resource surplus, progress rhythm, uncompleted task priority, external policy and natural environment changes is comprehensively sorted out, and key environmental parameters that directly affect the execution of coping strategies are extracted. These parameters can accurately reflect the unique scenario of the current project, providing practical environmental basis for subsequent strategy fine-tuning and avoiding blind adjustments that deviate from the current project status.

[0106] A matching degree evaluation system of key environmental parameters and preferred case coping strategies is established to evaluate the compatibility of strategy elements and environmental requirements from dimensions such as resources, time, scene, and standards, accurately identifying adaptation differences such as resource demand exceeding current reserves and execution time conflicting with progress. By clearly identifying the differences, a clear direction is established for subsequent element-level optimization and adjustment, avoiding missing key adaptation issues during optimization.

[0107] Based on the adaptation differences, the coping strategies are optimized at the element level, with specific adjustment schemes developed for different differences, such as optimizing resource allocation ratios, adjusting execution time nodes, and modifying implementation details. These optimization contents are refined into environmental adaptation factors. Each factor accurately corresponds to an adaptation difference, ensuring that strategy adjustments can directly address the core issues and avoid incomplete adjustments caused by general optimization.

[0108] In a dynamic context environment, the expected execution effect of the environment adaptation factor is verified, whether the adaptation difference can be solved, whether new problems are caused, whether the strategy performance is improved after the implementation of the factor, and effective adaptation factors meeting the performance threshold are screened out. This step can eliminate invalid or risky factors, ensure that the factors integrated into the strategy can effectively improve the strategy adaptability and execution effect, and avoid resource waste caused by invalid optimization.

[0109] The effective adaptation factor is integrated into the original coping strategy, and the corresponding element module is integrated according to the original logic and structure of the strategy, so that the factor and the strategy are seamlessly connected and cooperated. The finally formed enhanced coping strategy not only retains the core advantages of the optimized case strategy, but also adapts to the current project environment through the adaptation factor, greatly improves the actual execution performance of the strategy, and provides more accurate and efficient solutions for solving the current project deviation.

[0110] S6, the adaptive early warning and decision support information is transmitted to the landscape planning terminal, and the pre-stored project plan digital benchmark is driven to be visually updated.

[0111] In the embodiment of the present application, the adaptive early warning and decision support information is transmitted to the landscape planning terminal, and the pre-stored project plan digital benchmark is driven to be visually updated, comprising: According to the preset information format, the adaptive early warning and decision support information is packaged to obtain the standardized early warning and decision data package of the adaptive early warning and decision support information; The standardized early warning and decision data package is transmitted to the landscape planning terminal; The landscape planning terminal analyzes the standardized early warning and decision data package, and extracts the early warning content and decision suggestion information in the standardized early warning and decision data package; According to the early warning content and decision suggestion information, the pre-stored project plan digital benchmark is driven to be visually refreshed.

[0112] When the adaptive early warning and decision support information is packaged according to the preset information format to obtain the standardized early warning and decision data package of the adaptive early warning and decision support information, the preset information format is a fixed structure recognizable by the landscape planning terminal, including core fields such as early warning type, risk level, decision suggestion item, implementation priority, and associated entity identifier.

[0113] The early warning content in the adaptive early warning and decision support information is classified and filled into the corresponding field according to the type, the decision support information is decomposed into specific executable suggestion items, the implementation priority of each suggestion and the associated project entity are marked. The information is format checked to ensure that the content of each field meets the format requirements and there is no information missing or format disorder, and finally a standardized early warning and decision data package with standard structure and complete content is formed.

[0114] When transmitting the standardized early warning decision data packet to the landscape planning terminal, the security transmission channel preset by the landscape planning system is adopted, which has data encryption and integrity verification functions. After starting the transmission instruction, the system automatically establishes a connection with the landscape planning terminal, verifies the legality of the terminal identity, and ensures the accuracy of the data transmission object. The standardized early warning decision data packet is sent in segments according to the transmission protocol, and each segment of data is attached with a verification identifier. The receiving end verifies the data integrity in real time, and if data loss or damage occurs, the retransmission mechanism is triggered immediately. Until the data packet is completely transmitted to the landscape planning terminal, the transmission confirmation feedback is generated after the transmission is completed, ensuring that the data has been successfully delivered.

[0115] When the landscape planning terminal analyzes the standardized early warning decision data packet and extracts the early warning content and decision suggestion information in the standardized early warning decision data packet, the landscape planning terminal starts the built-in analysis program and reversely disassembles the standardized early warning decision data packet according to the preset information format.

[0116] First, verify the integrity and format validity of the data packet. After confirming that there is no exception, extract the information in each field one by one, sort and organize the early warning content according to the risk level, and clearly define the impact range and urgency of different levels of early warning. Classify the decision suggestion information according to the implementation priority, extract the core execution points and associated entities of each suggestion, remove the format identifiers and verification information in the data packet, and only keep the intuitive and usable early warning content and decision suggestion information to provide clear data support for subsequent visualization refresh.

[0117] According to the early warning content and decision suggestion information, drive the pre-stored project plan digital benchmark to refresh visually. The landscape planning terminal maps the extracted early warning content to the corresponding entity position of the pre-stored project plan digital benchmark, and uses different colors to identify different risk levels of early warning, such as using eye-catching colors for high-risk early warning and using corresponding gradient colors for low-risk early warning.

[0118] Combined with the decision suggestion information, add suggestion execution identifiers in the digital benchmark, such as annotating decision suggestion entries next to the task nodes that need to be adjusted, and using arrows or highlight boxes to prompt the association. Update the progress status, resource allocation, and other visual elements in the digital benchmark to make the early warning information and decision suggestions visually integrated with the project plan, ensuring that users can quickly capture risk points and response directions through the visual interface, and complete the visualization refresh of the pre-stored project plan digital benchmark.

[0119] The beneficial effects are that adaptive early warning and decision support information is encapsulated according to a preset format, with information such as early warning type, risk level, and decision recommendations filled into fixed fields to form a standardized early warning decision data package. This data package has a unified structure and regular content, avoiding information loss or misreading due to formatting issues during transmission, ensuring that the terminal can accurately identify the information, and laying a standardized foundation for subsequent analysis and application.

[0120] Standardized early warning decision data packets are transmitted to the landscape planning terminal through a pre-set secure transmission channel. During transmission, identity verification, data validation, and retransmission mechanisms ensure that the data packets arrive at the terminal completely, accurately, and without delay. This avoids information delays or omissions due to transmission failures, allowing staff to obtain early warning and decision-making information in a timely manner, thus gaining time for schedule adjustments.

[0121] When parsing standardized early warning decision data packages at the landscape planning terminal, the data is decomposed in reverse according to a preset format to extract early warning content sorted by risk level and decision recommendations categorized by priority, while removing redundant format markers. The extracted core information is intuitive and clear, allowing staff to quickly grasp key content without dealing with complex data formats, significantly improving the efficiency of information reading and use.

[0122] Based on the warning content and decision-making suggestions, the pre-stored project plan digital baseline is visually refreshed. Different colors are used to mark risk levels, add decision-making suggestion icons, and update progress status, so that warning information and adjustment suggestions are directly integrated into the plan baseline. Staff can intuitively see deviation risk points and corresponding countermeasures through the visual interface, without having to repeatedly compare information with the plan, quickly identify key points for progress adjustments, and improve the intuitiveness and efficiency of progress control.

[0123] like Figure 2 The diagram shown is a functional block diagram of a landscape planning progress tracking system provided in an embodiment of the present invention.

[0124] The landscape planning progress tracking system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the landscape planning progress tracking system 100 may include a data acquisition tagging module 101, a data cleaning and mapping module 102, a difference identification and calculation module 103, a deviation analysis report module 104, a decision support generation module 105, and an information synchronization and update module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0125] In this embodiment, the functions of each module / unit are as follows: The data collection tag module 101 is used for collecting a multi-modal original data set in a landscape planning process, and encapsulating a time stamp and a geographic coordinate tag on the multi-modal original data set, so as to obtain a spatio-temporal tagged original data package of the multi-modal original data set. The data cleaning graph module 102 is used for eliminating noise and inconsistency in the spatio-temporal tagged original data package, taking an entity in the spatio-temporal tagged original data package as a node, and taking a spatio-temporal and logical correlation relationship between entities as an edge, so as to construct a project progress state graph of the spatio-temporal tagged original data package. The difference identification calculation module 103 is used for identifying an attribute difference value of the project progress state graph and a pre-stored project plan digital benchmark. The deviation analysis report module 104 is used for mapping the attribute difference value to a historical multi-layer progress deviation identification rule library, judging a root source and an influence level of deviation, and generating a structured deviation analysis report of the project progress state graph. The decision support generation module 105 is used for generating adaptive early warning and decision support information of the landscape planning process by matching the structured deviation analysis report and a case feature in a historical coping strategy knowledge base. The information synchronization update module 106 is used for conveying the adaptive early warning and decision support information to a landscape planning terminal, and synchronously driving the pre-stored project plan digital benchmark to perform visual update.

[0126] In several embodiments provided in the present application, it should be understood that the disclosed method and system can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and another division mode can be used in actual implementation.

[0127] The modules described as separate components can or can not be physically separated, and the components displayed as modules can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. According to actual needs, part or all of the modules can be selected to achieve the purpose of the embodiment scheme.

[0128] In addition, each functional module in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of hardware plus software functional modules.

[0129] It is apparent for a person skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, but that the present application can be implemented in other concrete forms without departing from the spirit or essential characteristics of the present application.

[0130] The embodiment of the present application can acquire and process related data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology and application system for using digital computers or computer-controlled machines to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A method for tracking the progress of landscape planning, characterized in that, The method includes: S1. Collect multimodal raw datasets during the landscape planning process, and encapsulate the multimodal raw datasets with timestamps and geographic coordinate labels to obtain spatiotemporally labeled raw data packets of the multimodal raw datasets; S2. Eliminate noise and inconsistencies in the spatiotemporally tagged original data packet, and construct a project progress status map of the spatiotemporally tagged original data packet using entities in the spatiotemporally tagged original data packet as nodes and spatiotemporal and logical relationships between entities as edges. S3. Identify the attribute differences between the project progress status map and the pre-stored project plan digital benchmark; S4. Map the attribute difference values ​​to the historical multi-layer schedule deviation identification rule base, determine the root source and impact level of the deviation, and generate a structured deviation analysis report of the project schedule status map. S5. By matching the structured deviation analysis report with the case features in the historical response strategy knowledge base, adaptive early warning and decision support information for the landscape planning process is generated; S6. The adaptive early warning and decision support information is transmitted to the landscape planning terminal, and the pre-stored project plan digital benchmark is simultaneously driven to perform a visual update.

2. The landscape planning progress tracking method as described in claim 1, characterized in that, The process involves collecting multimodal raw datasets from the landscape planning process and encapsulating these datasets with timestamps and geographic coordinate labels to obtain a spatiotemporally labeled raw data package of the multimodal raw datasets, including: Image data, text description data, and numerical measurement data of the landscape planning area are integrated into a multimodal raw dataset of the landscape planning area. Eliminate the data format differences in the original multimodal dataset to obtain a standardized multimodal dataset for the landscape planning area; Add timestamps and geographic coordinate labels to the data items in the multimodal raw dataset to obtain the spatiotemporally labeled raw data package of the multimodal raw dataset.

3. The landscape planning progress tracking method as described in claim 1, characterized in that, The process of eliminating noise and inconsistencies in the spatiotemporally tagged original data packet, and constructing a project progress status graph of the spatiotemporally tagged original data packet using entities in the spatiotemporally tagged original data packet as nodes and spatiotemporal and logical relationships between entities as edges, includes: Repair missing, inconsistent, and logically contradictory data items in the spatiotemporally tagged original data packet to obtain a standardized data packet of the spatiotemporally tagged original data packet; The temporal and spatial features in the standardized data packet are combined using tensors to obtain the spatiotemporal feature vector of the standardized data packet. Based on the spatiotemporal feature vectors, calculate the spatiotemporal correlation matrix between entities in the standardized data packet; Analyze the task dependencies between the entities and establish the logical association matrix of the standardized data packets; The spatiotemporal correlation matrix and the logical correlation matrix are fused using multi-dimensional features to obtain the comprehensive correlation matrix of the standardized data packet; Using the entities as nodes and the element values ​​of the comprehensive association matrix as edge weights, a project progress status graph of the spatiotemporally tagged original data package is constructed.

4. The landscape planning progress tracking method as described in claim 3, characterized in that, The formula for calculating the spatiotemporal correlation strength in the spatiotemporal correlation matrix is ​​as follows: ； In the formula, For entities With entity The spatiotemporal correlation strength, Let be the standard deviation of the Gaussian distribution. It is a natural constant. As a weighting factor for the time dimension, Spatial dimension weighting factor, For entities With entity timestamp difference, For entities With entity The Euclidean distance between geographic coordinates For time scale parameters, This refers to the spatial scale parameter.

5. The landscape planning progress tracking method as described in claim 1, characterized in that, The identification of attribute differences between the project progress status map and the pre-stored project plan digital baseline includes: Extract the current state attribute set of entities in the project progress status map; Retrieve the set of planned status attributes for entities in the pre-stored project plan digital baseline; The current state attribute set is compared with the planned state attribute set on an entity-by-entity basis to identify a list of entities whose state attributes are inconsistent in the landscape planning process. Based on the entities in the list of entities with inconsistent status attributes, determine the attribute difference value between the current status attribute and the planned status attribute.

6. The landscape planning progress tracking method as described in claim 1, characterized in that, The process of mapping the attribute difference values ​​to a historical multi-layer schedule deviation identification rule base, determining the root cause and impact level of the deviation, and generating a structured deviation analysis report of the project schedule status map includes: Parse the entity identifier and deviation type information in the attribute difference values; The attribute difference values ​​are matched with the root cause determination rules in the historical multi-level progress deviation identification rule base to identify the root cause of the deviation. Based on the identified root cause, the impact level classification rules in the historical multi-level schedule deviation identification rule base are queried to determine the impact level of the deviation; The root causes and impact levels are integrated into a structured deviation analysis report of the project schedule status map.

7. The landscape planning progress tracking method as described in claim 1, characterized in that, The process of generating adaptive early warning and decision support information for the landscape planning process by matching the structured deviation analysis report with case features in the historical response strategy knowledge base includes: Deep semantic parsing is performed on the structured deviation analysis report to obtain the multidimensional feature vector of the structured deviation analysis report; In the historical response strategy knowledge base, historical cases with the highest contextual fit with the multidimensional feature vector are matched to generate a high-quality case candidate set for the landscape planning process; The timeliness, sustainability, and resource adaptability of the cases in the candidate set of high-quality cases are evaluated, and the preferred cases that meet the requirements are selected from the candidate set of high-quality cases. Based on the dynamic context of the current landscape planning project, the performance of the response strategy in the preferred case is fine-tuned to obtain the enhanced response strategy of the preferred case. By integrating the enhanced response strategies with real-time risk assessment results, adaptive early warning and decision support information for the landscape planning process is generated.

8. The landscape planning progress tracking method as described in claim 7, characterized in that, Based on the dynamic context of the current landscape planning project, the performance of the response strategy in the preferred case is fine-tuned to obtain an enhanced response strategy for the preferred case, including: Analyze the dynamic context of the current landscape planning project and extract its key environmental parameters. Establish a matching degree evaluation system between the key environmental parameters and the coping strategies in the preferred cases, and identify the adaptation differences between the strategy elements and environmental requirements in the matching degree evaluation system; Based on the adaptation differences, the response strategy is optimized and adjusted at the element level to obtain the environmental adaptation factor of the response strategy. Verify the expected performance of the environment adaptation factor in the dynamic context environment, and filter out the effective adaptation factors that meet the performance threshold. The effective adaptation factors are integrated into the response strategy to obtain the enhanced response strategy for the preferred case.

9. A method for tracking the progress of landscape planning as described in claim 1, characterized in that, The step of transmitting the adaptive early warning and decision support information to the landscape planning terminal and simultaneously driving the pre-stored project plan digital benchmark to perform visual updates includes: The adaptive early warning and decision support information is encapsulated according to a preset information format to obtain a standardized early warning and decision support data package. The standardized early warning decision data package is transmitted to the landscape planning terminal; The standardized early warning decision data package is parsed at the landscape planning terminal, and the early warning content and decision suggestion information in the standardized early warning decision data package are extracted. Based on the warning content and decision-making suggestions, the pre-stored project plan digital benchmark is driven to be visually refreshed.

10. A landscape planning progress tracking system for implementing the landscape planning progress tracking method of claim 1, the system comprising: The data acquisition and labeling module is used to collect multimodal raw datasets during the landscape planning process, and encapsulate the multimodal raw datasets with timestamps and geographic coordinate labels to obtain spatiotemporally labeled raw data packets of the multimodal raw datasets. The data cleaning and mapping module is used to eliminate noise and inconsistencies in the spatiotemporally labeled raw data package, and to construct a project progress status map of the spatiotemporally labeled raw data package using entities in the spatiotemporally labeled raw data package as nodes and spatiotemporal and logical relationships between entities as edges. The difference identification and calculation module is used to identify the attribute difference values ​​between the project progress status map and the pre-stored project plan digital benchmark; The deviation analysis report module is used to map the attribute difference values ​​to the historical multi-level schedule deviation identification rule base, determine the root source and impact level of the deviation, and generate a structured deviation analysis report of the project schedule status map. The decision support generation module is used to generate adaptive early warning and decision support information for the landscape planning process by matching the structured deviation analysis report with case features in the historical response strategy knowledge base. The information synchronization and update module is used to transmit the adaptive early warning and decision support information to the landscape planning terminal and synchronously drive the pre-stored project plan digital benchmark to perform visual updates.

Images