CN117555978B - Intelligent determining method for geographic model input data space range - Google Patents

Abstract

The invention provides an intelligent determining method for the space range of input data of a geographic model, which comprises the following steps: determining the characteristics of a geographic model to be constructed and the data type of input data; formulating identifiable knowledge rules based on the characteristics of the geographic model and the data type of the input data; based on the preset identifiable knowledge rule, determining the spatial range of input data of the geographic model to be constructed by combining a heuristic modeling method; and judging whether the data source meets the content and the space range requirements of the input data or not based on the determined space range, if not, iterating the reasoning process until the input data cannot be derived by other models or the data can meet the input conditions, and if so, obtaining the workflow of the geographic model to be calculated, wherein the workflow is configured with accurate space range input. The invention solves the problems that the prior art can cause incorrect spatial range of input data in the geographic modeling process, thereby causing linkage effect and generating incorrect geographic modeling result.

Description

An intelligent method for determining the spatial range of geographical model input data

Technical field

The invention relates to the technical field of geographical model construction, and in particular to an intelligent method for determining the spatial range of geographical model input data.

Background technique

As a critical step in the geographic modeling process, preparation of input data plays a vital role in ensuring successful execution of the modeling and obtaining complete and accurate results. The process includes not only preparing the appropriate content for each input, but also the appropriate spatial extent. Since the spatial extent of the input data depends on the characteristics of the adopted model and the corresponding input data type, the appropriate spatial extent of the input data may often be inconsistent with the spatial extent of the output target region of interest to the user. Especially when coupling multiple geographic models into a workflow, the various inputs in the workflow can be more complex and cumbersome when properly prepared. Effective methods are urgently needed to reduce the burden of input data preparation in the modeling process.

Depending on the modeling paradigm adopted, current input data preparation methods can be divided into two categories, namely program-oriented methods and goal-oriented methods. In a procedural-oriented approach, the user prepares input data starting from searching for raw data and establishes custom workflows to derive the data required for the model, often manually based on the modeler's knowledge of the specific modeling requirements. The goal-oriented approach aims to start from the user's modeling goals, automatically select appropriate input data and appropriate derived data models, and extend the model workflow in an iterative manner, thereby minimizing dependence on the user's modeling knowledge and skills. Thereby mitigating the drawbacks of the program-oriented approach. The basic strategy of goal-oriented methods is to formalize the knowledge required for input data preparation and apply it to advanced geological processing methods.

However, existing goal-oriented input data preparation methods focus on preparing appropriate data content (semantics and types) for the model, while ignoring the issue of appropriate spatial extent of the model input data. It is often up to the user to decide and set the appropriate spatial range of the input data. The user's research area is used as the spatial range of the input data for the model to be built (or a combination of models as a workflow model), even though it is possible to provide output covering the study area. The accuracy of its results cannot be guaranteed. This often leads to improper spatial range of input data during the geographic modeling process, which triggers a chain effect and produces incorrect geographic modeling results.

Contents of the invention

In order to overcome the deficiencies of the prior art, the purpose of the present invention is to provide an intelligent method for determining the spatial range of input data for a geographical model. The present invention solves the problem of inappropriate spatial range of input data that occurs in the process of geographical modeling in the prior art. situation, causing a knock-on effect and producing incorrect geographic modeling results.

In order to achieve the above objects, the present invention provides the following solutions:

An intelligent method for determining the spatial range of geographical model input data, including:

Determine the characteristics of the geographic model to be built and the data type of the input data;

Develop identifiable knowledge rules based on characteristics of the geographic model and data types of input data;

Based on preset identifiable knowledge rules and combined with heuristic modeling methods, determine the spatial range of the input data to be constructed for the geographical model;

Based on the determined spatial range, it is judged whether the data source meets the content and spatial range requirements of the input data. If not, the inference process is iterated until the input data cannot be derived by other models or there is data that can meet the input conditions. If so, the configuration is accurately obtained. Workflow of geographic model to be calculated with spatial extent input.

Preferably, it also includes:

Make a correction judgment on the spatial range of the input data of the above-mentioned geographical model to be calculated. If the required spatial range of the current data is inconsistent with the spatial range output by the upstream model, then the spatial range of the current input data is determined based on the current The spatial range of the input data is clipped and corrected to obtain a geographical model to be constructed that configures the accurate spatial range of the input data.

Preferably, the characteristics of the geographical model include:

Specific spatial extent requirements, connectivity expansion requirements, buffer distance expansion requirements, and spatial extent requirements to maintain the area of interest.

Preferably, the categories of input data include:

Point data, line data, area data and raster data.

Preferably, determining preset identifiable knowledge rules based on characteristics of the geographical model and categories of input data includes:

Classify and summarize according to the data type of the input data and the characteristics of the geographical model to be calculated, and obtain different classification combinations;

The need to determine the range of data space that appears in scenarios of different classification combinations based on classification induction;

Set the extraction process according to different spatial range requirements;

According to the classified and summarized data types and the model characteristics of the geographical model to be constructed, the model corresponds to different spatial range requirements and its extraction process, and identifiable knowledge rules are formulated.

Preferably, determining the data extraction process according to the classification combination includes:

The process of determining the first spatial range is to perform a spatial search in the integrated data set to determine whether there is a spatial range of the existing data set that can satisfy the first spatial range. If it is missing, determine whether the first spatial range can be obtained through the integration method;

The process of determining the second spatial range is to conduct a global connectivity search for data without directionality, and conduct a connectivity search by direction for data with directionality to determine the second spatial range;

Determine the third spatial range process. Based on the water flow direction calculated by DEM, the upstream catchment area is extracted and traced according to the water flow direction to obtain the complete watershed boundary. The extraction process distinguishes the water catchment area based on the upper reaches of the river and the water catchment area based on the upstream slope and determines third spatial range;

Determine the fourth spatial range process. The fourth spatial range is related to point elements. Thiessen polygons are constructed using point data. The smallest Thiessen polygon that intersects with the interest area is determined as the fourth spatial range process;

The process of determining the fifth space range is to directly expand outward to determine the fifth space range by specifying the buffer distance;

The process of determining the sixth spatial range, the sixth spatial range is a raster data type, used for terrain analysis, and the spatial range extending by one pixel grid size is determined as the sixth spatial range;

The process of determining the seventh spatial range, the seventh spatial range directly maintains the spatial range of the original area of interest.

According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

The invention provides an intelligent method for determining the spatial range of geographical model input data, which includes: determining the characteristics of the geographical model and the data type of the input data; and formulating identifiable knowledge based on the characteristics of the geographical model and the data type of the input data. Rules; based on preset identifiable knowledge rules, combined with heuristic modeling methods, to determine the spatial range of the input data to be constructed for the geographical model; based on the determined spatial range, determine whether the data source meets the content and spatial range requirements of the input data , if not, then iterate the inference process until the input data cannot be derived by other models or there is data that can meet the input conditions. If so, then the workflow of the geographic model to be calculated that is configured with accurate spatial range input is obtained. The present invention solves the problem in the prior art that improper spatial range of input data may occur during the geographical modeling process, thereby causing chain effects and producing incorrect geographical modeling results.

Description of the drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the drawings needed to be used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some of the drawings of the present invention. Embodiments, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without exerting creative efforts.

Figure 1 is a flow chart of a method for intelligently determining the spatial range of geographical model input data provided by an embodiment of the present invention;

Figure 2 is a schematic diagram of the principle of a method for intelligently determining the spatial range of geographical model input data provided by an embodiment of the present invention;

Figure 3 is a schematic diagram of the spatial range type provided by the embodiment of the present invention;

Figure 4 is a schematic flowchart of identifiable knowledge rules provided by an embodiment of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

The purpose of the present invention is to provide an intelligent method for determining the spatial range of geographical model input data. The present invention solves the problem of improper input data spatial range in the prior art during the geographical modeling process, thereby triggering chain effects and producing Issue with incorrect geographic modeling results.

In order to make the above objects, features and advantages of the present invention more obvious and understandable, the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

As shown in Figure 1, the present invention provides an intelligent method for determining the spatial range of geographical model input data, including:

Step 100: Determine the characteristics of the geographical model to be built and the data type of the input data;

Step 200: Develop identifiable knowledge rules based on the characteristics of the geographical model and the data type of the input data;

Step 300: Based on the preset identifiable knowledge rules and combined with the heuristic modeling method, determine the spatial range of the input data of the geographical model to be constructed;

Step 400: Based on the determined spatial range, determine whether the data source meets the content and spatial range requirements of the input data. If not, then iterate the inference process until the input data cannot be derived by other models or there is data that can meet the input conditions. If so, then get Configure the workflow of the geographic model to be calculated with accurate spatial range input.

This invention summarizes the characteristics of most geographical models and input data, and then forms a relatively general rule base. When the rules are actually applied, the geographical model to be constructed is reasoned.

Specifically, the present invention is based on a set of predefined knowledge rules and combined with heuristic modeling technology to automatically determine the distribution of all input data in the model workflow when performing geographic modeling in a research area of any spatial range. The spatial range is shown in Figure 2.

Specifically, in the geographical modeling process, the knowledge based on determining the appropriate spatial extent of the geographical model input data mainly considers two factors: the characteristics of the geographical model and the type of the corresponding input data. Knowledge about the impact of these two factors on the spatial extent of model input data can be systematically summarized according to model categories and data types, and a set of knowledge rules can be formed.

Knowledge classification related to model input data. According to the demand characteristics of the geographical model for the spatial range of the input data, it can be divided into four categories: specific spatial range requirements, connectivity expansion requirements, buffer distance expansion requirements, and maintaining the spatial range of the interest area. The types of model input data can be divided into point, line, area and raster data according to common types, and other special ones are classified into a separate category. The range of data space in different scenarios can be determined under the two classification combinations. details as follows:

In the first model classification, specific spatial extents covering the target area are specified, including administrative boundaries, watershed boundaries, or custom spatial extents. Regardless of whether the input data type corresponds to points, lines, polygons, or rasters, the spatial extent of the input data is consistent with that specific spatial extent.

The second type of model defines spatial extent through generalized connectivity. Generalized connectivity includes the connectivity of linear features such as roads and rivers, as well as the connectivity of data-derived information such as flow directions obtained from digital elevation models (DEMs). Therefore, this category only applies when the input data types are line and raster, since the other two data categories are inherently consistent with the target area.

The third type of model requires extending the buffer according to the target area. This is broken down into two subcategories: predefined distances and distances calculated based on context (like traditional buffer analysis). The contextually calculated distance is defined as the spatial extent required to buffer the analysis outward from the target domain; the predetermined distance varies depending on whether the input data is point or raster data. For point data, the spatial extent is defined as the smallest Thiessen polygon extent composed of points that covers the target domain. For raster data, the neighborhood window size is expanded based on the area of interest as the spatial extent.

The last four categories of models will maintain the spatial extent of the region of interest. This requires defining the spatial extent as the extent where the input data intersects the spatial extent of the area of interest. When the data types in the determination module fall outside the defined classifications, the target area extent will remain unchanged except that the model belongs to a specific spatial extent category. All spatial range types from A to E were finally determined, as shown in Figure 3.

Further, determining preset identifiable knowledge rules based on the characteristics of the geographical model and the category of the input data includes:

Classify and summarize according to the data type of the input data and the characteristics of the geographical model to be calculated, and obtain different classification combinations;

The need to determine the range of data space that appears in scenarios of different classification combinations based on classification induction;

Set the extraction process according to different spatial range requirements;

According to the classified and summarized data types and the model characteristics of the geographical model to be constructed, the model corresponds to different spatial range requirements and its extraction process, and identifiable knowledge rules are formulated.

Further, determining the data extraction process according to the classification combination includes:

Determining the data extraction process according to the classification combination includes:

The process of determining the first spatial range is to perform a spatial search in the integrated data set to determine whether there is a spatial range of the existing data set that can satisfy the first spatial range. If it is missing, determine whether the first spatial range can be obtained through the integration method;

The process of determining the second spatial range is to conduct a global connectivity search for data without directionality, and conduct a connectivity search by direction for data with directionality to determine the second spatial range;

Determine the third spatial range process. Based on the water flow direction calculated by DEM, the upstream catchment area is extracted and traced according to the water flow direction to obtain the complete watershed boundary. The extraction process distinguishes the water catchment area based on the upper reaches of the river and the water catchment area based on the upstream slope and determines third spatial range;

Determine the fourth spatial range process. The fourth spatial range is related to point elements. Thiessen polygons are constructed using point data. The smallest Thiessen polygon that intersects with the interest area is determined as the fourth spatial range process;

The process of determining the fifth space range is to directly expand outward to determine the fifth space range by specifying the buffer distance;

The process of determining the sixth spatial range, the sixth spatial range is a raster data type, used for terrain analysis, and the spatial range extending by one pixel grid size is determined as the sixth spatial range;

The process of determining the seventh spatial range, the seventh spatial range directly maintains the spatial range of the original area of interest.

Specifically, formulate knowledge rules that can be recognized by computers. Using knowledge rules in automated geographic model building. To simplify the automatic specification and extraction of spatial extents, we designed an 'if-then' rule. This 'if<condition>then<action>' rule utilizes the model and input data category as the "if condition" in the decision module. Subsequently, the spatial extent type we defined customized the appropriate extraction process as the corresponding "then action". details as follows:

For type A spatial range, first perform a spatial search in the integrated data set to determine whether there is existing data that meets the specific spatial range. If it is missing, it will be judged whether the spatial extent can be obtained through integrated methods (such as delimiting watershed boundaries).

Class B space range performs directional search on line data with display directionality, and performs global connectivity search on data without directionality.

Class C spatial range considers the flow direction calculated by DEM. The extraction of watershed boundaries distinguishes the catchment area based on the upper reaches of the river channel and the catchment area based on the upper reaches of the slope.

Class D spatial extent is related to point elements. Thiessen polygons are constructed using point data. The Thiessen polygons that intersect the interest area are determined as Class D spatial extents.

Class E directly expands outward to determine the required spatial range by specifying the buffer distance.

Class F is for raster data types, usually used for terrain analysis. For example, models such as slope calculation. In this case, expand outward by one cell.

Class G directly maintains the spatial extent of the original area of interest.

The 'if <condition> then <action>' rule is shown in Figure 4.

Specifically, knowledge rules are used in the process of model construction to determine the spatial extent of each input data. To ensure the practical performance of the developed method, especially in complex modeling workflows involving multiple models and inputs, our method is combined with heuristic modeling as follows:

After the user determines the area of interest for modeling and the model used, this method will gradually reason about the input data it requires. A basic problem unit consisting of modeling interest area, model, and data can be formed. According to the relevant model categories and data type categories we have predefined, the spatial range of the required input data can be determined under the rules. After determination, the designed method will determine whether the source data meets the content and spatial range requirements of the input data. If the requirements are not met, the inference process continues until the data cannot be derived by other models or there is data that meets the input conditions. The spatial extent of the current module will serve as the target area extent of subsequent modules, ensuring the integrity of each process calculation in the workflow.

A workflow model often has multiple branches. When encountering a situation where the model relies on multiple inputs, the method will generate different basic problem units based on different inputs to determine the spatial range respectively; when multiple models rely on the same input, the method will wait for the determination of the spatial range of the same data. and merge before iterative inference.

Furthermore, it also includes:

Make a correction judgment on the spatial range of the input data of the above-mentioned geographical model to be calculated. If the required spatial range of the current data is inconsistent with the spatial range output by the upstream model, then the spatial range of the current input data is determined based on the current The spatial range of the input data is clipped and corrected to obtain a geographical model to be constructed that configures the accurate spatial range of the input data.

Specifically, the model workflow for execution is rebuilt based on the determined input data spatial range. In order to ensure the integrity of the calculation results, the spatial range of the input data is often larger than the modeling area of interest, resulting in redundant calculations. For this reason, this method will determine whether each input needs to be spatially ranged before executing the workflow. Correction. The conditions for judgment are: If the spatial range of the current data has significantly changed compared to the input area of interest range, indicating that the range of the calculation result exceeds the spatial range of the downstream model input, the input data will be evaluated according to the corresponding spatial range in the downstream judgment module. Perform a cutting operation. Once all model input data in the workflow have been traversed and corrected, computational redundancy is effectively reduced.

The beneficial effects of the present invention are as follows:

(1) Automatic determination of the spatial range of all input data to the model (workflow). After the user selects a model and performs modeling in any selected area of interest, this method can automatically provide the user with feedback on the input data required by the model and the appropriate spatial range required for the intermediate data. This ensures the integrity of the input data, and its spatial range accuracy is based on the determined knowledge system and knowledge rules.

(2) Improvement of the accuracy of (intermediate) results. Compared with traditional model input data prepared directly based on the spatial range of the interest area, geographic model calculation based on the spatial range of the input data determined by this method can ensure that the model obtains complete results. The accuracy of the results is applied to the automated model building process based on determined knowledge rules.

(3) Reduce computational redundancy. In order to ensure the integrity of the calculation results, the spatial range of the input data is often larger than the requirements of the calculation results. During the calculation, this method will trim the data after confirming that the calculation results are accurate to avoid redundant calculations in the calculation.

Each embodiment in this specification is described in a progressive manner. Each embodiment focuses on its differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other.

This article uses specific examples to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method and the core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the present invention There will be changes in the specific implementation methods and application scope of the ideas. In summary, the contents of this description should not be construed as limitations of the present invention.

Claims (4)

1. An intelligent determining method for a geographic model input data space range is characterized by comprising the following steps:

determining the characteristics of a geographic model to be constructed and the data type of input data;

formulating identifiable knowledge rules based on the characteristics of the geographic model and the data type of the input data;

based on a preset identifiable knowledge rule and combined with a heuristic modeling method, determining the spatial range of input data of a geographic model to be constructed;

judging whether the data source meets the content and space range requirements of the input data or not based on the determined space range, if not, iterating the reasoning process until the input data cannot be derived by other models or the data can meet the input conditions, and if so, obtaining the workflow of the geographic model to be calculated with the accurate space range input configured;

the determining a preset identifiable knowledge rule based on the characteristics of the geographic model and the category of the input data comprises the following steps:

classifying and summarizing according to the data types of the input data and the characteristics of the geographic model to be calculated to obtain different classification combinations;

determining the requirements of the data space range appearing under the scenes of different classification combinations according to classification induction;

setting an extraction flow according to different space range requirements;

formulating identifiable knowledge rules according to the classified and induced data types, different space range requirements corresponding to the model characteristics of the geographic model to be constructed and the extraction flow thereof;

setting an extraction flow according to different space range requirements, including:

determining a first space range flow, performing space search in the integrated data set to determine whether the space range of the existing data set can meet the first space range, and if the space range of the existing data set is missing, judging whether the first space range can be obtained by an integration method;

determining a second space range flow, performing global connectivity search on the data without directivity, and performing connectivity search on the data with directivity according to the direction to determine a second space range;

determining a third space range flow, extracting and tracing an upstream water collecting area according to the water flow direction based on the water flow direction calculated by the DEM to obtain a complete river basin boundary, and determining a third space range by distinguishing the water collecting area based on the upstream of the river channel from the water collecting area based on the upstream of the slope in the extracting process;

determining a fourth spatial range flow, the fourth spatial range being related to the point element, constructing a Thiessen polygon using the point data, the minimum Thiessen polygon intersecting the region of interest being determined as the fourth spatial range flow;

determining a fifth space range flow, and directly expanding outwards to determine a fifth space range through a designated buffer distance;

determining a sixth spatial range procedure, wherein the sixth spatial range is used for terrain analysis on the type of raster data, and the spatial range of the raster size of one pixel which is externally expanded is determined as a sixth spatial range;

and determining a seventh spatial range flow, wherein the seventh spatial range directly maintains the spatial range of the original interest region.

2. The method for intelligently determining the spatial extent of input data of a geographic model according to claim 1, further comprising:

and carrying out correction judgment on the spatial range of the input data of the geographic model to be calculated, and if the required spatial range of the current data is inconsistent with the spatial range output by the upstream model, carrying out clipping correction on the spatial range of the current input data based on the spatial range of the current input data to obtain the geographic model to be constructed for configuring the accurate spatial range of the input data.

3. The method for intelligently determining the spatial extent of input data of a geographic model according to claim 1, wherein the features of the geographic model comprise:

a specific spatial range requirement, a connectivity expansion requirement, a buffer distance expansion requirement, and a spatial range requirement to maintain a region of interest.

4. The method for intelligently determining the spatial extent of input data of a geographic model according to claim 1, wherein the category of the input data comprises:

dot data, line data, face data, and raster data.