Landscape area intelligent planning and designing method based on data analysis

Abstract

The invention relates to the technical field of landscape area planning design and data analysis, and discloses an intelligent landscape area planning design method based on data analysis. The method comprises the steps of collecting multi-source data of a landscape area, analyzing and extracting elements such as terrain elevation, vegetation coverage, water body distribution and a building outline, converting the elements into a standard format and establishing association, and generating a landscape basic data set; and querying a planning rule base by using the data set, matching landscape types and planning constraint rule entries, calculating a rule application complexity value, judging a data coincidence state, and performing single planning compliance evaluation. A plurality of landscape basic data sets are integrated, shared entities among different planning projects are found, the association strength level among the projects is calculated, and a multi-project association graph is constructed; and calculating a risk value based on the multi-project association graph and single-project planning compliance assessment, and generating a comprehensive planning risk prompt. Through data analysis, the scientificity and compliance of landscape area planning and design and the collaboration among multiple projects are improved.

Description

Technical Field

[0001] This invention relates to the field of landscape area planning and design and data analysis technology, specifically a data analysis-based intelligent planning and design method for landscape areas. Background Technology

[0002] In the past, landscape planning and design work relied heavily on the experience and subjective judgment of planners. Traditional methods often had limitations when analyzing elements such as topographic elevation, vegetation cover, water distribution, and building outlines within a landscape area. For example, when analyzing topographic elevation, simple surveying data was often used, making it difficult to comprehensively and meticulously represent the complex characteristics of the terrain. Subtle topographic undulations that significantly impact the landscape effect were easily overlooked, potentially leading to irrational subsequent landscape layouts. For instance, planning large event spaces in areas with steep slopes not only increases construction costs but also poses safety hazards.

[0003] In vegetation cover analysis, traditional methods rely primarily on manual field surveys, which are inefficient and their accuracy is greatly affected by human factors. For large landscape areas, it is difficult to quickly obtain comprehensive information on vegetation species, distribution range, and growth status, making it impossible to provide timely and accurate data support for landscape planning. This may result in plant configuration schemes lacking scientific rigor, failing to fully consider the ecological habits and landscape combination effects of different plants, leading to poor vegetation growth or poor landscape harmony.

[0004] Traditional analytical methods for water body distribution do not adequately consider the dynamic changes of water bodies. They only understand the static location and extent of water bodies, while ignoring factors such as seasonal changes in water level, direction and speed of water flow. This may lead to problems such as flooding of landscape facilities and poor drainage when carrying out landscape construction around water bodies.

[0005] In analyzing building silhouette elements, traditional methods struggle to effectively integrate the relationship between buildings and the surrounding landscape. Focusing solely on the building's appearance and function without considering its impact on the skyline and spatial hierarchy from a holistic landscape perspective can easily lead to a disconnect between the building and the landscape, thus damaging the overall aesthetic appeal of the landscape.

[0006] In terms of planning compliance assessment, traditional methods lack a systematic and comprehensive evaluation framework. They mainly rely on planners' memory and understanding of relevant laws and regulations, and determine whether the planning scheme is compliant through manual comparison and judgment. This method is highly subjective, inefficient, and prone to overlooking important compliance issues, increasing the risk and cost of rectification in the later stages of the project.

[0007] Traditional landscape planning also falls short in considering the interrelationships between multiple projects. It often treats each project in isolation, failing to fully recognize potential connections such as shared topographical features, interconnected water bodies, and vegetation. This makes it impossible to optimize resource allocation and achieve coordinated development when planning multiple projects, resulting in wasted landscape resources and inconsistent landscape effects. For example, two adjacent parks, without considering shared water resources during planning, may have their own distinct water feature designs, leading to vastly different styles and disrupting the overall landscape continuity.

[0008] With the rapid development of information technology, we have entered the era of big data, generating and recording massive amounts of data. In the field of landscape planning, this data covers multiple aspects such as topography, climate, population distribution, vegetation growth, and urban traffic, providing a wealth of information sources for landscape planning. Simultaneously, the continuous emergence and development of new technologies such as Geographic Information Systems (GIS), remote sensing, Global Positioning Systems (GPS), big data analytics, and artificial intelligence algorithms have made the collection, processing, and analysis of this massive amount of data possible. For example, GIS technology can accurately collect and analyze spatial data such as topography, geomorphology, and hydrology of landscape areas; RS technology can quickly obtain information on vegetation cover and water body distribution over large areas; and big data analytics can extract potential information related to landscape planning from massive amounts of urban population and traffic data, such as the needs and preferences of different population groups for landscape functions and the impact of traffic flow around the landscape on its accessibility. However, most landscape planning and design units currently fail to fully utilize these new technologies and big data resources, still employing traditional planning methods. Therefore, there is an urgent practical need to develop a data analysis-based intelligent planning and design method for landscape areas. It can give full play to the advantages of big data and new technologies, solve many problems existing in traditional landscape planning, and improve the scientificity, rationality and efficiency of landscape planning. Summary of the Invention

[0009] The purpose of this invention is to provide a data analysis-based intelligent planning and design method for landscape areas to solve the problems mentioned in the background art.

[0010] To achieve the above objectives, the present invention provides a smart planning and design method for landscape areas based on data analysis, the method comprising: By collecting multi-source data of the landscape area, we analyze and extract topographic elevation, vegetation cover, water body distribution and building outline elements, convert the extracted elements into standard data format and establish the relationship between the elements to generate a basic landscape dataset. Using the landscape basic dataset, the rule entries that match the landscape type and planning constraints are queried from the planning rule base. The rule application complexity value is calculated based on the number of matching entries and the logical level. Based on the rule application complexity value, the landscape data is compared with the rule limit value, effective time period and geographical scope item by item to determine the data compliance status and generate a single planning compliance assessment. By integrating multiple landscape basic datasets, shared topographic features, associated water bodies, and vegetation entities are discovered among different planning projects. The association strength level between projects is calculated based on the number and category of shared entities. Based on the association strength level between projects, a network structure with planning entities as nodes and shared relationships as edges is constructed to form a multi-project association graph. Based on the multi-project relationship diagram, risk values ​​are calculated by combining the compliance assessment of individual plans of related projects, and a comprehensive planning risk warning is generated.

[0011] Preferably, the specific steps for generating the landscape basic dataset include: simultaneously collecting topographic point cloud data, multispectral image data, and surface hydrological monitoring data through a sensor network and remote sensing equipment deployed in the landscape area to form multimodal landscape raw data; performing spatiotemporal alignment and coordinate system unification processing on the multimodal landscape raw data to eliminate scale differences and positional deviations during data acquisition; scanning the processed data using feature extraction algorithms to identify and separate contour features of topographic elevation, patch outline features of vegetation cover, boundary features of water body distribution, and geometric shape features of building outlines; vectorizing the identified features and mapping them to a preset standardized geographic information field format; performing topological relationship checks on the fields that have completed format conversion to establish spatial adjacency, inclusion, and intersection relationship identifiers between topographic, vegetation, water body, and building elements, and finally generating a landscape basic dataset with complete spatial attributes and relationships.

[0012] Preferably, obtaining the rule application complexity value includes: parsing the classification code field describing the landscape type and the keyword field of the planning constraint from the generated landscape basic dataset; semantically matching the classification code and keywords with the terminology dictionary in the planning rule base, performing semantic disambiguation and standardization replacement on ambiguous or polysemous field values ​​to form a standardized combination of landscape type and planning constraint descriptions; using this standardized combination as a query condition, performing a full-text index search in the planning rule base to find all relevant rule entries; performing structural analysis on each retrieved rule entry, and statistically analyzing its nesting depth of logical judgment conditions, the number of other dependent rule entries, the start and end time span of the rule's own validity period, the number of geographical boundary coordinates to which the rule applies, and the total number of times the rule has been called in historical projects; based on the statistical results, calculating a structural complexity weight for each rule entry, and weighted summing the structural complexity weights of all matching rule entries to finally obtain a quantified rule application complexity value.

[0013] Preferably, the step-by-step operation for generating a single-plan compliance assessment is as follows: based on the calculated rule application complexity value, determine the level of detail and priority order for comparing the landscape basic data with the rule restriction values; sequentially read the value or attribute of each data field from the landscape basic dataset and compare it with the corresponding restriction value in the active rule entry. The comparison operation includes checking whether the value is within the limit range, whether the attribute enumeration value conforms to the allowed list, and whether the data collection timestamp is within the rule validity period; for rules involving geographic spatial range, call the geographic information system component to perform spatial overlay analysis on the geographic coordinates or range polygon in the landscape data and the applicable geographical range specified in the rule clause, and calculate the overlapping area ratio of the two; summarize the comparison results and spatial analysis results of all fields, generate a clear pass or fail status mark for the compliance of each data field with each relevant rule, comprehensively determine the compliance status of the single plan based on all marks, and generate an assessment report containing detailed non-compliance items.

[0014] Preferably, the steps for obtaining the inter-project association strength level include: extracting geomorphic unit identifiers describing topographic features, watershed codes describing water bodies, and ecological zone numbers describing vegetation communities from the landscape basic datasets corresponding to multiple independent landscape planning projects; cross-comparing these identifiers and codes from different datasets to identify the same or similar entities that appear repeatedly in different projects; classifying the identified shared entities, counting the frequency of each type of entity in different projects, and recording the specific distribution location information of each entity in all appearing projects; pre-setting basic weights based on the importance of the entity category, and combining their frequency of occurrence and the dispersion of distribution, dynamically calculating a quantitative index reflecting the degree of close association between multiple projects through shared entities, i.e., the inter-project association strength level, through an association strength calculation model.

[0015] Preferably, the process of constructing the multi-project association graph includes: selecting a set of projects with an association strength exceeding a threshold based on the level of association strength between projects to construct the network graph; creating a unique node for each selected project and assigning attributes to each node, the content of which is derived from the core field summary of the landscape basic dataset of that project; establishing an edge between any two project nodes if they share a specific type and number of entities; the weight of the edge is determined by calculating the type weight and number of shared entities and the level of association strength between projects; and storing and organizing all nodes and weighted edges to construct a network graph structure that displays the complex relationships between multiple landscape planning projects, i.e., the multi-project association graph.

[0016] Preferably, the steps for generating comprehensive planning risk warnings are detailed as follows: In the constructed multi-project association graph, a graph traversal algorithm is used to explore all paths within a certain length range, starting from each planning project node; the node sequence traversed by each path is recorded, and the compliance assessment results of the individual planning corresponding to these nodes are queried; path characteristics are analyzed, including path length, connection density of the end nodes of the path, richness of data fields contained in the starting node of the path, and number of nodes passed through the path; combining path characteristics and compliance status of the passed nodes, a risk propagation model is used to calculate the potential risk impact value of each path on the starting node; the risk impact values ​​of all paths are aggregated to obtain the comprehensive risk score of each node; based on the preset risk level threshold range, the corresponding risk level is marked for each node and displayed visually in the graph structure, while generating a comprehensive planning risk warning document containing a list of high-risk nodes and a description of risk-related paths.

[0017] Preferably, the method further includes a step of adjusting the plan based on risk warnings: parsing the comprehensive planning risk warning document to identify landscape areas or planning elements marked as high-risk; for each high-risk item, tracing back its original data in the landscape basic dataset and verifying the specific rule clauses it triggers; matching a set of feasible adjustment schemes from the planning strategy library according to the constraint intent of the rule clauses and the characteristics of the original data; simulating the effect of each adjustment scheme on the landscape basic dataset and reassessing the compliance risks after modification; comparing the simulation results of each adjustment scheme, selecting the scheme that effectively reduces risks and has the least impact on planning objectives, and generating a detailed planning adjustment instruction set.

[0018] Preferably, the verification step after generating the planning adjustment instruction set includes: applying the planning adjustment instruction set to the original landscape basic dataset to generate a new version of the landscape basic dataset; using the new dataset, re-executing the rule matching and compliance assessment process to obtain new individual planning compliance assessment results; recalculating the risk scores of relevant projects based on the new assessment results and the original multi-project association diagram; checking whether all the new risk scores have dropped below the acceptable threshold, and verifying whether the adjustment process has introduced new potential risk conflicts.

[0019] Preferably, the method further includes: integrating the validated planning adjustment instruction set, the adjusted landscape basic dataset, and the compliance assessment results; and converting them into a complete planning and design package containing a floor plan, a section analysis diagram, a planning indicator description table, and a risk assessment appendix, in accordance with industry standard planning drawings and document specifications.

[0020] Compared with the prior art, the beneficial effects of the present invention are: This invention collects multi-source data from landscape areas, analyzes and extracts elements such as topographic elevation, vegetation cover, water distribution, and building outlines, converts them into standard data formats, establishes relationships between elements, and generates a basic landscape dataset. This process breaks through the limitations of traditional data collection and processing, enabling the integration and standardization of previously scattered and inconsistent data. Previously, data from different sources, such as topographic data from traditional surveying and vegetation data from manual records, varied in format and accuracy, making them difficult to use directly for comprehensive landscape analysis. The basic landscape dataset generated by this invention, however, is comprehensive and accurate, providing a solid and reliable data basis for subsequent landscape planning and design. Whether for small-scale community landscape planning or the design of large urban parks, planners can obtain detailed and accurate data from this dataset, gaining a clearer and more comprehensive understanding of the landscape area and laying a good foundation for developing scientific and reasonable planning schemes.

[0021] By utilizing a landscape dataset, this invention queries a planning rule base to match rule entries between landscape types and planning constraints. Based on the number of matching entries and their logical hierarchy, it calculates the rule application complexity value and determines the data's compliance status, generating a single-item planning compliance assessment. This approach changes the traditional model that relies on subjective human judgment of planning compliance, introducing a scientific quantitative analysis method. By systematically comparing landscape data with rule limits, effective time periods, and geographical scope, it can proactively and comprehensively identify potential non-compliance issues in the planning process. In traditional planning, planners may overlook certain legal provisions or misunderstand regulations, leading to compliance problems in the later stages of planning, requiring significant time and cost for rectification. This invention's single-item planning compliance assessment identifies these potential problems in the early stages of planning, helping planners adjust their planning direction in a timely manner, avoiding unreasonable planning, ensuring that the planning scheme complies with relevant laws and regulations from the outset, saving planning costs, and improving planning efficiency.

[0022] By integrating multiple landscape datasets, this invention can discover shared topographic features, associated water bodies, and vegetation entities among different planning projects. Based on the number and category of shared entities, the association strength between projects is calculated, and a network structure with planning entities as nodes and shared relationships as edges is constructed, forming a multi-project association graph. This process allows planners to clearly see the inherent connections between different projects, changing the previous approach of viewing each planning project in isolation. In urban landscape planning, multiple park, green space, and municipal infrastructure construction projects may coexist. Through the multi-project association graph, planners can discover shared resources such as water bodies and topography among these projects, thereby achieving resource sharing and optimized allocation during planning. For example, when two adjacent park plans share the same water body resource, a unified water landscape design can be implemented, avoiding redundant construction and resource waste. Simultaneously, it makes the landscape style of different projects more coordinated and unified, improving overall planning efficiency and effectiveness, and maximizing the utilization of landscape resources. Attached Figure Description

[0023] Figure 1 This is a schematic diagram illustrating the working principle of the intelligent planning and design method for landscape areas based on data analysis as described in this invention. Figure 2 A flowchart for generating the basic landscape dataset; Figure 3 A flowchart for generating a compliance assessment of a single plan. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Please see Figure 1 This invention provides a data analysis-based intelligent planning and design method for landscape areas. The method includes: collecting multi-source data from the landscape area, parsing and extracting topographic elevation, vegetation cover, water body distribution, and building outline elements; converting the extracted elements into a standard data format and establishing relationships between elements to generate a landscape basic dataset. Using the landscape basic dataset, querying a planning rule base to match rule entries between the landscape type and planning constraints; calculating the rule application complexity value based on the number of matching entries and logical hierarchy; comparing the landscape data with rule restrictions, effective time periods, and geographical scope item by item based on the rule application complexity value to determine data compliance status and generate a single-item planning compliance assessment. Integrating multiple landscape basic datasets to discover shared topographic features, associated water bodies, and vegetation entities among different planning projects; calculating the inter-project association strength level based on the number and category of shared entities; constructing a multi-project association graph based on the inter-project association strength level, with planning entities as nodes and shared relationships as edges; and calculating risk values ​​based on the multi-project association graph and the single-item planning compliance assessment of associated projects to generate a comprehensive planning risk warning.

[0026] Example 1: See Figure 2The generation of the landscape baseline dataset relies on a sensor network and remote sensing equipment deployed in the landscape area. These devices simultaneously collect topographic point cloud data, multispectral image data, and surface hydrological monitoring data to form multimodal landscape raw data. The multimodal landscape raw data undergoes spatiotemporal alignment and coordinate system normalization to eliminate scale differences and location biases during data acquisition. Feature extraction algorithms scan and identify the processed data, separating contour features of topographic elevation, patch outlines of vegetation cover, boundary features of water body distribution, and geometric features of building outlines. The sensor network consists of ground-based laser scanners and weather stations deployed at fixed locations. Remote sensing equipment includes UAV platforms equipped with multispectral cameras and high-resolution optical satellites. Topographic point cloud data records the three-dimensional coordinates of the land surface using lidar equipment in a pulse-echo manner. Multispectral image data captures the reflection characteristics of surface objects in different electromagnetic spectrum bands. Surface hydrological monitoring data compiles real-time readings of river water levels, soil moisture, and precipitation intensity. The spatiotemporal alignment process uses the GPS timing module to add a unified timestamp and geodetic coordinates to each frame of data. The coordinate system unification process transforms spatial data from different sources to the projection plane under the national geodetic coordinate system. The scale difference correction uses a resampling algorithm to interpolate low-resolution data onto a spatial grid consistent with high-resolution data. The position deviation correction uses control points for geometric registration to make images and point clouds from different sources accurately overlap in space.

[0027] Feature extraction algorithms scan and identify the processed data, separating contour features of terrain elevation, patch outlines of vegetation cover, boundary features of water body distribution, and geometric features of building outlines. Contour features of terrain elevation are extracted by constructing an irregular triangular network model from point cloud data. Patch outlines of vegetation cover are segmented using a normalized vegetation index threshold to identify green vegetation areas in multispectral images. Boundary features of water body distribution are analyzed by combining near-infrared water absorption characteristics with mountain shadow lines from a digital elevation model to eliminate mountain shadow interference. Geometric features of building outlines are identified using edge detection operators to pinpoint straight lines and right-angled structures in the image. All identified features are vectorized and mapped to a pre-defined standardized geographic information field format. The converted fields undergo topological relationship checks to establish spatial adjacency, inclusion, and intersection relationships between terrain, vegetation, water bodies, and building elements, ultimately generating a landscape dataset with complete spatial attributes and relationships. Vectorization transforms raster-style feature boundaries into vector polygons composed of nodes and line segments. Standardized geographic information field formats define the coordinate storage structure, attribute field names, and data types of point, line, and polygon features. Topological relationship checks verify whether there are overlaps, gaps, or unclosed boundary errors between polygons. Spatial adjacency records feature identifiers that share boundaries. Containment relationships label the level of features that are completely enclosed. Intersection relationships calculate the geometric intersection of overlapping areas of features.

[0028] The calculation of the rule application complexity value is performed by parsing the classification code field describing the landscape type and the keyword field of the planning constraint from the generated landscape base dataset. The classification code and keyword are semantically matched with the terminology dictionary in the planning rule base. For ambiguous or polysemous field values, semantic disambiguation and standardized replacement are performed to form a standardized combination of landscape type and planning constraint descriptions. The classification code field of the landscape type adopts the three-level coding system in the national land use classification standard. The keyword field of the planning constraint is extracted from the article summary of the local urban planning management technical regulations. The terminology dictionary constructs a mapping table from local colloquial names to standard terms, including synonyms and hyponyms. Semantic disambiguation analysis analyzes the contextual dependencies of words in sentences and selects the most relevant word meanings. Standardized replacement replaces non-standard expressions such as "small grove" with the standardized term "tree forest". The normalized combination is used as a query condition to perform a full-text index search in the planning rule base to find all relevant rule entries. For each retrieved rule entry, structural analysis is performed to statistically analyze its logical judgment condition nesting depth, the number of other dependent rule entries, the start and end time span of the rule's own validity period, the number of geographical boundary coordinates to which the rule applies, and the total number of times the rule has been called in historical projects. The full-text index search uses inverted index technology to quickly locate the rule text containing keywords. The structural analysis decomposes the rule condition statement into an abstract syntax tree and calculates the maximum depth of the condition branches. The number of other dependent rule entries is counted to determine the external rule numbers referenced in the rule text. The start and end time span of the rule's own validity period is calculated to determine the total number of days from the effective date to the expiration date. The number of geographical boundary coordinates to which the rule applies is counted to determine the number of polygon vertices defining the management boundary. The total number of times the rule has been called in historical projects is aggregated from the log records of the project approval database.

[0029] Based on statistical results, a structural complexity weight is calculated for each rule entry. The structural complexity weights of all matching rule entries are then weighted and summed to obtain a quantified value for the rule application complexity. The calculation of the structural complexity weight uses a multi-factor linear model, assigning different coefficients to nesting depth, number of dependencies, time span, number of boundary points, and number of calls. The weighted summation formula multiplies the weight of each rule entry by its matching score in the current query and then sums them. The output of the rule application complexity value is a dimensionless numerical value used to indicate the resource investment required for rule application. Higher nesting depth of rule entries indicates a more complex logical hierarchy of conditional judgments, requiring more computational steps for verification. A greater number of dependent rule entries means more external constraints need to be checked, increasing system load. A longer start and end time span of the rule's validity period may introduce temporal logic conflicts requiring special handling. A greater number of geographical boundary coordinates to which the rule applies leads to a greater computational load for spatial overlay analysis. The total number of times a rule has been called in historical projects reflects its practicality; frequently called rules often require optimization of execution efficiency. The standardized field structure of the landscape foundation dataset provides clear patterns for classification coding and keyword parsing. The terminology dictionary of the planning rule base maintains the consistency of professional terminology and avoids ambiguity in natural language. The structural parsing process transforms rule text into machine-measurable topological indicators. The design of structural complexity weights balances the impact of different factors on the difficulty of rule execution. The quantitative results of rule application complexity values ​​provide an objective basis for the fine-grained control of subsequent compliance assessments. The acquisition of multimodal landscape raw data integrates ground-based and airborne sensing platforms to achieve multi-dimensional data coverage. Spatiotemporal alignment and coordinate unification processing establish a spatial benchmark for data fusion. Feature extraction algorithms adapt to the morphological characteristics of different types of geographic elements. Vectorization transformation and topology checking ensure the logical consistency of geographic data. Semantic matching and disambiguation processing bridge the gap between natural language and standardized terminology. The rule retrieval and parsing mechanism realizes the transformation from text rules to computable indicators. The weight calculation model transforms qualitative rule features into quantitative complexity measures.

[0030] Example 2: See Figure 3The compliance assessment of a single planning item is based on the calculated rule application complexity value, which determines the level of detail and priority for comparing the landscape basic data with the rule restriction values. The level of rule application complexity directly determines the level of detail in the data comparison performed by the system. Higher values ​​trigger a deep verification mode to perform multiple checks on each data field, while lower values ​​use a standard verification mode for basic compliance checks. The priority order is arranged according to the sensitivity of the constraint factors in the rule entries. Structural constraints involving ecological protection red lines and public safety have the highest priority, while guiding constraints involving landscape aesthetics and functional layout are processed later. The system reads the values ​​or attributes of each data field from the landscape basic dataset and compares them with the corresponding restriction values ​​in the active rule entries. The data fields are read in groups according to element categories: the terrain elevation field reads the altitude and slope values; the vegetation cover field reads the vegetation type code and coverage percentage; the water body distribution field reads the water area and water quality level; and the building outline field reads the building area and floor height. Activated rule entries are obtained by querying the set of rules in the planning rule base that match the validity period field with the current timestamp. Rule restriction values ​​include numerical thresholds, enumerated lists, and spatiotemporal ranges.

[0031] The comparison operation includes checking whether the values ​​are within the specified range, whether the attribute enumeration values ​​conform to the allowed list, and whether the data collection timestamp is within the rule's validity period. The value range check uses an interval comparison algorithm to verify whether the data value falls within the minimum and maximum value range defined by the rule. The attribute enumeration value check matches whether the data attribute value exists in the set of allowed values ​​defined by the rule through a lookup table. The data collection timestamp check compares the chronological relationship between the data acquisition time and the rule's effective and expiration periods. For rules involving geospatial extent, the Geographic Information System (GIS) component is invoked to perform spatial overlay analysis between the geographic coordinates or extent polygons in the landscape data and the applicable geographical area specified in the rule clauses, calculating the ratio of their overlapping area. The GIS component calls the intersection operation method of the spatial analysis engine; the geographic coordinates in the landscape data are converted into boundary polygons, and the applicable geographical area in the rule clauses is extracted from the rule base as a vector boundary. The overlapping area ratio is obtained by calculating the ratio of the geometric intersection area of ​​the two polygons to the area of ​​the landscape data polygon.

[0032] The comparison results and spatial analysis results of all fields are summarized to generate a clear pass or fail status flag for each data field's compliance with each relevant rule. The status flag uses a Boolean value: true for pass and false for fail. The comparison results and spatial analysis results of each data field are recorded in the compliance check log. Based on all flags, the compliance status of the individual plan is comprehensively determined, and an assessment report containing detailed non-compliance items is generated. The compliance status determination adopts a weighted voting mechanism, with the failure flag of key constraint fields having veto power. When the number of failure flags of general constraint fields reaches a certain threshold, it is determined as an overall non-compliance. The assessment report structure includes a project overview, a compliance summary, detailed inspection records, and a non-compliance item list. The non-compliance item list specifies the identifier, rule clause number, actual value and limitation requirements, degree of deviation, and improvement suggestions for each failure field.

[0033] The correlation strength level between projects was obtained by extracting geomorphic unit identifiers describing topographic features, watershed codes describing water bodies, and ecological zone numbers describing vegetation communities from the landscape base datasets corresponding to multiple independent landscape planning projects. Geomorphic unit identifiers adopted the geomorphic type code from the National Geographic Information Classification Code, watershed codes adopted the watershed unit code from the water resources zoning coding system, and ecological zone numbers adopted the ecological region number from the ecological functional zoning. The extraction process used database queries to filter the unique value list of corresponding fields from the attribute table of each project's landscape base dataset. These identifier codes from different datasets were cross-referenced to identify identical or similar entities that appeared repeatedly in different projects. The cross-reference used a hash table data structure to store the identifier set of all projects. By traversing the set and counting the number of projects in which each identifier appeared, identifiers appearing more than once were identified as shared entities. Identification of similar entities allowed for partial matching of codes; for example, sub-watershed codes within the same watershed with a common prefix were considered related entities.

[0034] The identified shared entities are categorized and their frequency of occurrence across different projects is statistically analyzed, and the specific distribution location information of each entity across all participating projects is recorded. Shared entities are categorized into three main types: geomorphic units, watersheds, and vegetation communities. Each category is further subdivided according to the entity coding level. Frequency is calculated by counting the number of times each entity code appears in the project set. Distribution location information is extracted from the spatial coordinate field of the landscape dataset using the entity's geometric center point coordinates or boundary coordinates. Based on the entity's category importance, a pre-defined basic weight is assigned. Combined with its frequency of occurrence and the dispersion of its distribution, a correlation strength calculation model dynamically calculates a quantitative indicator reflecting the degree of association between multiple projects through shared entities—the inter-project association strength level. The basic weights for category importance are set by domain experts, with geomorphic entities having a higher weight than vegetation entities, and watershed entities having the highest weight. Frequency of occurrence is directly multiplied by the basic weight as a positive factor. The dispersion of distribution is measured by calculating the standard deviation of the entity's location coordinates across all participating projects; a larger standard deviation indicates a higher degree of dispersion. The correlation strength calculation model substitutes the basic weight, frequency of occurrence, and degree of dispersion into a linear combination formula. The degree of dispersion is used in the calculation in the form of the reciprocal to reflect the characteristic of clustered distribution strengthening the correlation. The final output of the correlation strength level between items is a normalized numerical value between zero and one.

[0035] The calculation of inter-project association strength relies on a standardized identifier system of the landscape baseline dataset, ensuring comparability of entities across different projects. Cross-comparison algorithms rapidly identify shared resources through efficient set operations. The association strength calculation model comprehensively considers the essential importance of entities, their actual degree of sharing, and spatial distribution characteristics. The quantified inter-project association strength provides a precise linking basis for subsequent construction of multi-project association graphs. The strength value directly determines the existence and weight of edges between project nodes, thus affecting the topological characteristics of the network structure and risk propagation paths. Example 3: The formation of the multi-project association graph is based on the association strength level between projects. A set of projects with association strength exceeding a threshold is selected for constructing the network graph. The association strength level is a quantitative numerical indicator reflecting the degree of connection between different landscape planning projects through shared entities. The threshold is set using a statistical distribution method, calculating the average of all project association strength levels plus a standard deviation. Project pairs with strength values ​​higher than this threshold are considered to have significant association. The selection process iterates through each row and column of the project association strength level matrix, adding the identifiers of projects that meet the threshold condition to the project set for the network graph. A unique node is created for each selected project, and attributes are assigned to each node. The attribute content originates from the core field summary of the project's landscape basic dataset. Node creation assigns a globally unique identifier using a random generation algorithm based on UUID version 4. Node attributes extract key descriptive fields from the landscape basic dataset, including project name, planned area, main terrain type, dominant vegetation category, water area ratio, and building density index. The attribute summary is extracted from the original dataset using a data transformation script and converted into a key-value pair format, stored in the node data structure.

[0036] An edge is established between any two project nodes if they share a specific type and number of entities. The types of shared entities include geomorphic units (topographic features), watershed units (water bodies), and vegetation community units. The shared quantity is determined by the requirement that entities of the same type appear simultaneously in the landscape datasets of both projects. A minimum shared quantity constraint is set for edge establishment, such as sharing at least one geomorphic unit and one watershed unit, or sharing more than three vegetation community units. The edge weight is determined by calculating the type weight and quantity of shared entities, as well as the inter-project association strength level. The type weight of shared entities is predefined based on the importance of the entity category, with geomorphic units having the highest weight, followed by watershed units, and vegetation community units having the lowest weight. The inter-project association strength level is used as a moderating factor in the calculation, and the shared entity quantity acts as a multiplier to amplify the influence of the base weights. The formula for calculating the edge weight is as follows:

[0037] All nodes and weighted edges are stored and organized to construct a network graph structure, namely the multi-project association graph, which displays the complex relationships between multiple landscape planning projects. Node storage uses an adjacency list data structure to record the identifier, attribute dictionary, and list of connecting edges for each node. Weighted edges store the target node identifier and weight value. The network graph structure supports graph traversal operations and topology analysis, and the adjacency list implementation makes node queries and edge traversal highly efficient. The constructed multi-project association graph is a weighted undirected graph where nodes represent planning projects, edges represent project associations, and weights represent association strength. The generation of comprehensive planning risk warnings uses a graph traversal algorithm to explore all paths within a certain length range starting from each planning project node in the constructed multi-project association graph. The graph traversal algorithm adopts a breadth-first search strategy, visiting adjacent nodes layer by layer from the starting node, with the path length limited to three steps to avoid combinatorial explosion. During path exploration, the sequence of nodes traversed by each path is recorded, and the compliance assessment results of the corresponding individual planning projects are queried. The node sequence is stored as an ordered list recording the access order from the starting node to the ending node. The compliance assessment results for each single plan are retrieved from the assessment database, with scores ranging from zero to one; higher scores indicate better compliance. Path characteristics are analyzed, including path length, the density of connections at the ending nodes, the richness of data fields contained in the starting node, and the number of nodes traversed along the path. Path length is calculated as the number of hops, i.e., the number of edges in the path. The density of connections at the ending nodes is measured by calculating the degree centrality index of the ending nodes, which equals the number of connected edges divided by the maximum possible number of connections in the network graph. The richness of data fields contained in the starting node is calculated by counting the number of key-value pairs in the starting node's attribute dictionary. The number of nodes traversed along the path is calculated by counting the number of intermediate nodes in the path sequence excluding the starting and ending nodes. Combining path characteristics and the compliance status of traversed nodes, a risk propagation model is used to calculate the potential risk impact value of each path on the starting node. The risk propagation model defines a risk transmission function; low compliance scores at traversed nodes propagate risk along the path to the starting node, with path characteristics acting as a moderating factor affecting the propagation intensity. The longer the path, the greater the risk attenuation; the higher the density of connections at the end nodes, the greater the risk concentration; the richer the data fields at the starting node, the stronger the risk buffering capacity; and the more nodes along the path, the more obvious the risk accumulation effect.

[0038] The risk impact values ​​of all paths are aggregated to obtain the comprehensive risk score for each node. The aggregation operation uses the maximum value aggregation method to select the value with the largest risk impact on the starting node among all paths as the node's comprehensive risk score. Based on a preset risk level threshold range, each node is labeled with a corresponding risk level and visually highlighted in the graph structure. A comprehensive planning risk warning document is generated, containing a list of high-risk nodes and descriptions of risk-related paths. The risk level threshold range is divided into three levels: 0 to 0.3 (low risk, marked green), 0.3 to 0.7 (medium risk, marked yellow), and 0.7 to 1 (high risk, marked red). Visual highlighting in the network diagram interface uses color coding for nodes: high-risk nodes are displayed in red, medium-risk nodes in yellow, and low-risk nodes in green. The comprehensive planning risk warning document uses a structured text format. The list of high-risk nodes is arranged in descending order of risk score. Each node includes a project identifier and risk value. The risk-related path descriptions detail the node sequence and compliance status of each high-risk path. The construction of a multi-project association graph integrates discrete planning projects into an interconnected system. The edge weight calculation formula quantifies the strength dimension of the association between projects, the graph traversal algorithm reveals potential risk transmission chains, and the risk propagation model transforms local compliance issues into global risk assessments. The output of comprehensive planning risk alerts provides planning managers with risk visualization and decision support at the project cluster level. The identification of high-risk nodes helps to prioritize the allocation of review resources, and the description of risk association paths indicates the key channels for risk diffusion.

[0039] Example 4: Steps for Planning Adjustment Based on Risk Alerts The comprehensive planning risk alert document identifies landscape areas or planning elements marked as high-risk. This document is a structured electronic file containing a list of high-risk nodes and a description of risk-related paths. The parsing process uses a Document Object Model (DOM) interface to read the document content and a pattern matching algorithm to locate entries marked with high-risk levels. Each entry corresponds to a planning project or a specific geographic element within a project. For each high-risk item, the original data in the landscape dataset is traced back to verify the specific rule clauses it triggers. The tracing operation executes a query in the relational database of the landscape dataset based on the unique identifier of the high-risk item, extracting all field records corresponding to that identifier. Rule clause verification accesses the planning rule base by linking the rule number associated with the high-risk item, retrieving the complete rule text, including constraints, limits, and applicable instructions.

[0040] Based on the binding intent of the rules and the characteristics of the raw data, a set of feasible adjustment schemes are matched from the planning strategy library. The binding intent of the rules is analyzed to extract core limiting objectives from the rule text, such as protecting ecologically sensitive areas and controlling development intensity. The characteristics of the raw data are analyzed to assess the direction and magnitude of the gap between the current data state and the rule requirements. The planning strategy library is a database storing standard adjustment measures, including fields such as scheme number, applicable rule type, adjustment action description, and expected impact scope. The matching process calculates the fit score between the scheme and the problem based on the rule type and the data gap, selecting the top-scoring schemes to form a candidate adjustment scheme set. The modification effect of each adjustment scheme on the landscape basic dataset is simulated, and the compliance risks after modification are reassessed. The simulation involves creating a temporary copy of the landscape basic dataset and applying data change operations such as modifying attribute values, moving feature positions, and adding or deleting feature records according to the steps described in the adjustment scheme. The reassessment process calls the single-plan compliance assessment workflow to check the rule compliance of the modified temporary dataset, generating a new compliance status score and risk level. By comparing the simulation results of various adjustment schemes, the scheme that effectively reduces risk and has the least impact on planning objectives is selected, generating a detailed planning adjustment instruction set. An evaluation matrix is ​​established to compare the degree of risk reduction and the degree of deviation from planning objectives for each scheme. The degree of deviation from planning objectives is measured by calculating the changes in key planning indicators before and after the adjustment. The planning adjustment instruction set is an executable operation sequence file containing instruction numbers, operation objects, action types, and parameter values.

[0041] The planning adjustment instruction set is applied to the original landscape baseline dataset to generate a new, adjusted landscape baseline dataset. The application process executes each instruction in the planning adjustment instruction set through a script interpreter, directly modifying the database records of the original landscape baseline dataset. The rule matching and compliance assessment process is re-executed using the new dataset to obtain new individual planning compliance assessment results. This process completely replicates the initial assessment steps to ensure consistency in assessment standards. Based on the new assessment results and the original multi-project association diagram, the risk scores of relevant projects are recalculated. It is checked whether all new risk scores have fallen below the acceptable threshold and whether the adjustment process has introduced any new potential risk conflicts. The risk score recalculation uses the original multi-project association diagram structure, updating only the compliance status score in the node attributes, and re-runs the risk propagation model and aggregation algorithm. The acceptable threshold is preset to a fixed value in the system configuration. New potential risk conflicts are checked by comparing the differences in risk warning documents before and after the adjustment to identify newly added high-risk items. See Table 1, which shows the matching logic for some adjustment schemes in the planning strategy library.

[0042] Table 1: Matching Table for Adjustment Schemes in the Planning Strategy Library

[0043] The verification steps following the planning adjustment instruction set apply the instruction set to the original landscape basic dataset, generating a new, adjusted landscape basic dataset. The application process is atomic, ensuring that all instructions either execute successfully or are rolled back to maintain data integrity. The new landscape basic dataset inherits the structure and metadata of the original dataset; only content fields are changed according to the instructions. The rule matching and compliance assessment processes are re-executed, using the same rule engine and assessment algorithm to ensure comparability. New individual planning compliance assessment results are stored side-by-side with the original assessment results for easy difference analysis. Multi-project correlation diagrams remain static, with only node attributes updated to avoid interference from changes in network structure. Risk scores are recalculated using the original model parameters to ensure consistent scoring standards. Acceptable threshold checks are performed, traversing all relevant nodes to confirm that risk scores are below the upper limit. New potential risk conflicts are verified by fully scanning the new comprehensive planning risk warning documents to check for high-risk items not present in the original documents. The planning adjustment steps based on risk warnings achieve closed-loop management from risk identification to measure implementation. Predefined schemes in the planning strategy library provide standardized response strategies, and simulation execution and reassessment constitute a reliable decision support mechanism. The verification steps following the planning adjustment instruction set ensure the effectiveness and safety of the adjustment measures, preventing risk transfer or the emergence of derivative problems. The entire process embodies the design philosophy of dynamic adjustment and iterative optimization, enabling landscape planning to adapt to changes in constraints and continuously improve. The executable format of the planning adjustment instruction set facilitates automated system processing, and the rigor of the verification steps meets the reliability requirements of engineering practice.

[0044] Example 5: Integrating the validated planning adjustment instruction set, the adjusted landscape baseline dataset, and the compliance assessment results, and converting them into a complete planning and design package according to industry-standard planning drawings and document specifications, including a floor plan, cross-sectional analysis diagram, planning indicator description table, and risk assessment appendix. The planning adjustment instruction set records all change operation steps from the original scheme to the final scheme, the adjusted landscape baseline dataset stores the latest spatial coordinates and attribute information of all landscape elements, and the compliance assessment results provide detailed status markers for each planning element's compliance with relevant rules.

[0045] Taking a specific urban waterfront landscape planning project as an example, the planning adjustment instruction set contains fifteen specific operational instructions, covering aspects such as reshaping pedestrian walkways, expanding wetland planting areas, and reducing service building heights. The adjusted landscape dataset correspondingly updated the geometry and attribute tables of vector elements, rearranged the node coordinates of pedestrian walkway elements, expanded the polygonal boundaries of wetland planting areas, and corrected the height field value of service building elements from twelve meters to nine meters. The compliance assessment results showed that the previously violated rule of "waterfront building height not exceeding ten meters" was now marked as compliant, and the rule of "wetland protection area not less than five hectares" also met the standard requirements. The integration process used a data assembly pipeline to convert and reorganize the formats of these three types of input materials. The planning adjustment instruction set was parsed into a version change log, the adjusted landscape dataset was rendered into graphic elements, and the compliance assessment results were summarized into statistical indicators.

[0046] The generation of the plan layout map extracts the spatial coordinate information of all point, line, and area elements from the adjusted landscape dataset, assigning corresponding legend symbols and color fill schemes according to the element classification codes. Pedestrian walkways are represented by continuous brown solid lines, wetland vegetation areas are distinguished by light green mesh fill patterns, and service building elements are marked with gray rectangular symbols to indicate building outlines. The map layout follows drafting standards, placing the north arrow, scale bar, and legend frame in standard positions. Map annotations use standardized fonts to label element names and key dimensions. The creation of the profile analysis map requires first determining the position and direction of the profile lines on the plan layout map. Profile lines are typically arranged perpendicular to contour lines to fully reflect the topographic relief characteristics. The topographic elevation information in the adjusted landscape dataset is interpolated using a digital elevation model to calculate the elevation value of each point on the profile line. The generated elevation points are connected to form topographic profile lines, which are then overlaid to display the vertical distribution relationship of planning elements, such as the relationship between building height and topography.

[0047] The planning indicator description table extracts the actual and target values ​​of core planning control indicators from the compliance assessment results. The table rows correspond to different indicator categories, and the columns include the indicator name, unit of calculation, current value, planned value, and compliance status field. Indicator names include green space ratio, water surface ratio, building density, plot ratio, and height control value. Current values ​​are derived from the statistical summary of the adjusted landscape dataset, and planned values ​​are derived from the restrictions in the planning rule base. Compliance status is automatically determined by comparing current and planned values. The risk assessment appendix is ​​based on the latest comprehensive planning risk warning document, extracting a list of high-risk nodes and risk-related path descriptions, converting them into narrative text and diagrams. The high-risk node list indicates the project number, risk level, and risk score. The risk-related path description uses arrow diagrams to illustrate the risk transmission relationship and includes textual explanations.

[0048] The complete planning and design package outputs in DWG vector graphics files, PDF documents, and XLSX data tables. DWG files contain geometric data and layer information for plan layouts and sectional analysis diagrams; PDF files integrate the layout of all drawings, tables, and appendices; and XLSX files store the raw data for the planning indicator description table. The data conversion process uses the application programming interface (API) of computer-aided design software to generate drawings in batches, uses a document generation engine to assemble the various parts into a coherent document, and exports structured tables from a spreadsheet library. The complete planning and design package's file structure is organized by project number, version number, and date for easy version management and archiving. The integrated operation ensures the traceability of planning adjustment instructions, the visualization of the adjusted landscape baseline dataset, and the understandable presentation of compliance assessment results. Plan layouts reflect the spatial layout relationships of the planning scheme, sectional analysis diagrams reveal the vertical structure of the terrain and buildings, the planning indicator description table quantifies core planning indicators, and the risk assessment appendix warns of potential problems. The completion of the complete planning and design package marks the end of the intelligent planning and design process for the landscape area, providing an authoritative basis for planning approval and project implementation. The change history of the planning adjustment instruction set records the evolution of the scheme. The adjusted landscape basic dataset serves as a spatial basis to support further analysis. The embedding of compliance assessment results reflects the compliance-driven design concept.

[0049] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A data-driven intelligent planning and design method for landscape areas, characterized in that: The method includes the following steps: By collecting multi-source data of the landscape area, we analyze and extract topographic elevation, vegetation cover, water body distribution and building outline elements, convert the extracted elements into standard data format and establish the relationship between the elements to generate a basic landscape dataset. Using the landscape basic dataset, the rule entries that match the landscape type and planning constraints are queried from the planning rule base. The rule application complexity value is calculated based on the number of matching entries and the logical level. Based on the rule application complexity value, the landscape data is compared with the rule limit value, effective time period and geographical scope item by item to determine the data compliance status and generate a single planning compliance assessment. By integrating multiple landscape basic datasets, shared topographic features, associated water bodies, and vegetation entities are discovered among different planning projects. The association strength level between projects is calculated based on the number and category of shared entities. Based on the association strength level between projects, a network structure with planning entities as nodes and shared relationships as edges is constructed to form a multi-project association graph. Based on the multi-project relationship diagram, risk values ​​are calculated by combining the compliance assessment of individual plans of related projects, and a comprehensive planning risk warning is generated.

2. The intelligent planning and design method for landscape areas based on data analysis according to claim 1, characterized in that, The specific steps for generating the landscape basic dataset include: simultaneously collecting topographic point cloud data, multispectral image data, and surface hydrological monitoring data through sensor networks and remote sensing equipment deployed in the landscape area to form multimodal landscape raw data; performing spatiotemporal alignment and coordinate system unification processing on the multimodal landscape raw data to eliminate scale differences and positional deviations during data acquisition; scanning the processed data using feature extraction algorithms to identify and separate contour features of topographic elevation, patch outline features of vegetation cover, boundary features of water body distribution, and geometric shape features of building outlines; vectorizing the identified features and mapping them to a preset standardized geographic information field format; performing topological relationship checks on the fields that have completed format conversion to establish spatial adjacency, inclusion, and intersection relationship identifiers between topographic, vegetation, water body, and building elements, and finally generating a landscape basic dataset with complete spatial attributes and relationships.

3. The intelligent planning and design method for landscape areas based on data analysis according to claim 1, characterized in that, The process of obtaining the rule application complexity value includes: parsing the classification code field describing the landscape type and the keyword field of the planning constraint from the generated landscape basic dataset; semantically matching the classification code and keywords with the terminology dictionary in the planning rule base, performing semantic disambiguation and standardization replacement on ambiguous or polysemous field values ​​to form a standardized combination of landscape type and planning constraint descriptions; using this standardized combination as a query condition, performing a full-text index search in the planning rule base to find all relevant rule entries; performing structural analysis on each retrieved rule entry, and statistically analyzing its nesting depth of logical judgment conditions, the number of other dependent rule entries, the start and end time span of the rule's own validity period, the number of geographical boundary coordinates to which the rule applies, and the total number of times the rule has been called in historical projects; based on the statistical results, calculating a structural complexity weight for each rule entry, and weighted summing the structural complexity weights of all matching rule entries to finally obtain a quantified rule application complexity value.

4. The intelligent planning and design method for landscape areas based on data analysis according to claim 1, characterized in that, The step-by-step operation for generating a compliance assessment of a single plan is as follows: Based on the calculated rule application complexity value, determine the level of detail and priority order for comparing the landscape basic data with the rule restriction values; sequentially read the value or attribute of each data field from the landscape basic dataset and compare it with the corresponding restriction value in the active rule entry. The comparison operation includes checking whether the value is within the limit range, whether the attribute enumeration value conforms to the allowed list, and whether the data collection timestamp is within the rule validity period; for rules involving geographic spatial range, call the geographic information system component to perform spatial overlay analysis on the geographic coordinates or range polygon in the landscape data and the applicable geographical range specified in the rule clause, and calculate the overlapping area ratio of the two; summarize the comparison results and spatial analysis results of all fields, generate a clear pass or fail status mark for the compliance of each data field with each relevant rule, comprehensively determine the compliance status of the single plan based on all marks, and generate an assessment report containing detailed non-compliance items.

5. The intelligent planning and design method for landscape areas based on data analysis according to claim 1, characterized in that, The steps for obtaining the inter-project association strength level include: extracting geomorphic unit identifiers describing topographic features, watershed codes describing water bodies, and ecological zone numbers describing vegetation communities from the landscape basic datasets corresponding to multiple independent landscape planning projects; cross-comparing these identifiers and codes from different datasets to identify the same or similar entities that appear repeatedly in different projects; classifying the identified shared entities, counting the frequency of each type of entity in different projects, and recording the specific distribution location information of each entity in all appearing projects; pre-setting basic weights based on the importance of the entity category, and combining their frequency of occurrence and the dispersion of distribution, dynamically calculating a quantitative index reflecting the degree of close association between multiple projects through shared entities, i.e., the inter-project association strength level, through an association strength calculation model.

6. The intelligent planning and design method for landscape areas based on data analysis according to claim 1, characterized in that, The process of constructing the multi-project association graph includes: selecting a set of projects with an association strength exceeding a threshold based on the level of association strength between projects to construct the network graph; creating a unique node for each selected project and assigning attributes to each node, with the attribute content derived from the core field summary of the project's landscape basic dataset; establishing an edge between any two project nodes if they share a specific type and number of entities; determining the weight of the edge by calculating the type weight and number of shared entities and the level of association strength between projects; and storing and organizing all nodes and weighted edges to construct a network graph structure that displays the complex relationships between multiple landscape planning projects, i.e., the multi-project association graph.

7. The intelligent planning and design method for landscape areas based on data analysis according to claim 1, characterized in that, The steps for generating comprehensive planning risk warnings are detailed as follows: In the constructed multi-project association graph, a graph traversal algorithm is used to explore all paths with lengths within a certain range, starting from each planning project node; Record the sequence of nodes traversed by each path and query the compliance assessment results of the individual plans corresponding to these nodes; analyze path characteristics, including path length, connection density of the end nodes of the path, richness of data fields contained in the starting nodes of the path, and number of nodes passed through the path; combine path characteristics and compliance status of the nodes passed through, and use a risk propagation model to calculate the potential risk impact value of each path on the starting node. Aggregate the risk impact values ​​of all paths to obtain a comprehensive risk score for each node; Based on the preset risk level threshold range, each node is marked with a corresponding risk level and displayed visually in the graph structure. At the same time, a comprehensive planning risk warning document containing a list of high-risk nodes and a description of risk-related paths is generated.

8. The intelligent planning and design method for landscape areas based on data analysis according to claim 1, characterized in that, The method also includes a step of adjusting the plan based on risk warnings: parsing the comprehensive planning risk warning document to identify landscape areas or planning elements marked as high-risk; for each high-risk item, tracing back its original data in the landscape basic dataset and verifying the specific rule clauses it triggers; matching a set of feasible adjustment schemes from the planning strategy library according to the constraint intent of the rule clauses and the characteristics of the original data; simulating the effect of each adjustment scheme on the landscape basic dataset and reassessing the compliance risks after modification; comparing the simulation results of each adjustment scheme, selecting the scheme that effectively reduces risks and has the least impact on planning objectives, and generating a detailed planning adjustment instruction set.

9. The intelligent planning and design method for landscape areas based on data analysis according to claim 8, characterized in that, The verification steps after generating the planning adjustment instruction set include: applying the planning adjustment instruction set to the original landscape basic dataset to generate a new version of the landscape basic dataset; using the new dataset, re-executing the rule matching and compliance assessment process to obtain new individual planning compliance assessment results; recalculating the risk scores of relevant projects based on the new assessment results and the original multi-project association diagram; checking whether all the new risk scores have dropped below the acceptable threshold, and verifying whether the adjustment process has introduced new potential risk conflicts.

10. The intelligent planning and design method for landscape areas based on data analysis according to claim 9, characterized in that, The method also includes: integrating a validated set of planning adjustment instructions, an adjusted landscape baseline dataset, and compliance assessment results; and converting them into a complete planning and design package containing a floor plan, a section analysis diagram, a planning indicator description table, and a risk assessment appendix, in accordance with industry standard planning drawings and document specifications.

Images