Workflow-based surface live analysis self-adaptive scheduling method and device and application of workflow-based surface live analysis self-adaptive scheduling method and device

Abstract

The invention discloses a workflow-based surface live analysis self-adaptive scheduling method, device and application. The method comprises the following steps: step P1, defining a self-adaptive scheduling framework; p2, designing a self-adaptive identifier of the system; step P3, parameterizing the operation environment of the system; step P4, parameterizing the time of system analysis; step P5, analyzing the definition of observation data; step P6, background field definition is analyzed; and step P7, defining an analysis algorithm. The device comprises an adaptive scheduling framework definition module, a system adaptive identification design module, a system operation environment parameterization module, an analysis time parameterization module, an analysis observation data definition module, an analysis background field definition module and an analysis algorithm definition module. The workflow-based self-adaptive scheduling method for surface live analysis is applied to business scenes such as surface live analysis product development, and the automation level, the resource utilization efficiency and the capability of coping with complex scenes of a surface live analysis system can be effectively improved.

Description

Technical Field

[0001] This invention relates to the field of surface condition information processing and system construction technology. Specifically, it relates to a workflow-based adaptive scheduling method, apparatus, and application for surface condition analysis. Background Technology

[0002] Surface condition analysis integrates data resources from various sources and of various types, utilizing data fusion and assimilation technologies to construct a seamless, gridded forecasting service at "zero time." Most surface condition analysis systems consist of multiple functions forming an organic whole, including data acquisition and preprocessing, background field processing, data fusion analysis, and product post-processing. They also involve numerical calculations and process control, and currently, real-time operations typically employ workflow-based unified process scheduling and monitoring management.

[0003] During product development, quality assessment, business operations, and product backtesting, there are various changes such as the need for product production and comparative analysis with different development schemes or methods, the need for production analysis of data from different sources within the same category, the need to start production of products of different quality at different times, the need for product backtesting over historical periods, and the need for fault-tolerant and supplementary calculations during real-time operation. These changes lead to repeated development of product code and redundant system construction, posing certain challenges to product production scheduling. Therefore, developing a workflow-based adaptive scheduling method is crucial for ensuring the development and real-time production of products based on surface reality analysis. Summary of the Invention

[0004] Therefore, the technical problem to be solved by the present invention is to provide an adaptive scheduling method, device and application based on workflow-based surface reality analysis, so as to solve the problems of code redundancy and system duplication in traditional static scheduling.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] An adaptive scheduling method based on workflow-driven surface condition analysis includes the following steps:

[0007] Step P1: Define an adaptive scheduling framework based on the development content of the surface condition analysis product, and form a tree-structured definition content file corresponding to the surface condition analysis product; this provides the basic structure and framework for subsequent parameterization and adaptive scheduling.

[0008] Step P2: Based on the established adaptive scheduling framework, plan and design the system adaptive identifiers. The system adaptive identifiers include the surface reality analysis system identifier, business scenario identifier, application scheme identifier, and sequence number identifier.

[0009] Step P3: Parameterize the runtime environment of each system adaptive identifier and generate the corresponding runtime environment configuration file;

[0010] Step P4: Parameterize the analysis time of the surface condition analysis system from three aspects: analysis schedule, basic analysis cycle, and analysis time window;

[0011] Step P5: Define and classify the analytical observation data of the surface condition analysis system;

[0012] Step P6: Parameterize the background field type and background field selection strategy of the surface condition analysis system;

[0013] Step P7: Parameterize the analysis algorithm of the surface condition analysis system to achieve modularization and strategy-based analysis algorithm.

[0014] The adaptive scheduling method based on workflow-based surface condition analysis, in step P1, defines the adaptive scheduling framework as follows:

[0015] (P1-1) Based on the surface condition analysis, define the corresponding suite (work package) name and framework content for the product development content;

[0016] (P1-2) Define the family name and framework content for different scenarios within the suite (job package) node framework. These generally include rapid operation scenario, standard operation scenario, product rework scenario, historical product back-calculation scenario, and different sensitive test evaluation scenarios, etc.

[0017] (P1-3) Within the family (job family) node framework for dealing with different scenarios, the process is planned as observation data preprocessing, background field processing, fusion analysis and product postprocessing, etc., and the corresponding sub-family (sub-job family) content or task (task) elements are defined according to the processing logic.

[0018] (P1-4) Adopting a modular structural design, the functional modules involved in each scenario in the surface reality fusion analysis are extracted to complete the definition of the name and execution function framework of the task module for a specific function;

[0019] (P1-5) Classify the logically combinable task functional modules and use sub-family (sub-job family) for framework definition and representation.

[0020] In the aforementioned workflow-based adaptive scheduling method for surface condition analysis, step P2 involves the following identifiers: Surface condition analysis system identifier to distinguish different condition analysis systems; Business scenario identifier to distinguish different business scenarios such as Rapid Run (RAPID), Standard Run (CYCLE), Product Rework (RERUN), Historical Product Backtesting (HIST), and Sensitive Test (TC); Application scheme identifier to distinguish different surface condition analysis technical schemes; and Serial number identifier, which is optional and uses two digits, primarily used to identify different scheduling branches when process splitting and parallel processing occur under the same conditions (surface condition analysis system identifier, business scenario identifier, and application scheme identifier). The surface condition analysis system identifier, business scenario identifier, application scheme identifier, and serial number identifier are separated by an underscore "_".

[0021] By parsing the system identifier content, the adaptive scheduling configuration script can accurately match the scheduled running tasks with the corresponding environmental resources, data, background fields, fusion methods, etc. at the appropriate running time, thereby realizing process scheduling and adaptation.

[0022] In the above-mentioned adaptive scheduling method for surface reality analysis based on workflow, step P3 defines and declares the parameterized content of the system adaptive identifier's runtime environment in the configuration file. This includes the main directory for system script execution, the directory for observation data input and quality control processing, the directory for background field input and processing, the directory for fusion analysis product generation, and the directory for product post-processing. Depending on the business scenario, each adaptive identifier can use its own independent parameterized configuration definition content or share the same parameterized configuration definition content.

[0023] The above-mentioned adaptive scheduling method for surface condition analysis based on workflow addresses the relatively fixed time period of the surface condition analysis product. Parameterizing the analysis time intervals aims to enable the system to flexibly and accurately respond to the operational needs of different time scales and triggering conditions. In step P4, the method for customizing the analysis time interval is to define the analysis time interval using a Crontab-like expression to achieve automatic processing of the product analysis cycle. For example, it defines that the current time interval's condition analysis will be performed every 3 hours with a 45-minute delay.

[0024] The method for defining the basic analysis cycle is as follows: set the basic analysis time interval, for example, start it once every 20 minutes, and determine the specific time of surface condition analysis based on product efficiency, observation data, background field data, and analysis of product generation.

[0025] Dynamic adjustment of analysis time windows: By parameterizing the start and end times of analysis time windows, the analysis time windows can be adjusted according to the actual needs of the business scenario, thereby starting the required real-time analysis tasks at the appropriate time and improving the timeliness of the analysis.

[0026] In the aforementioned workflow-based adaptive scheduling method for surface condition analysis, step P5 involves the observation data, which is a crucial input for condition analysis. Refining its definition and classification aims to enable the system to adaptively identify, select, and preprocess observation data from multiple sources. This is especially important when accessing observation data of the same type but from different sources, requiring targeted data definition and preprocessing configuration. This invention defines and classifies the observation data from the following dimensions:

[0027] Data source type: Different identifiers are designed for observation data from different data sources to distinguish them, and these identifiers are defined parametrically.

[0028] Prioritization strategy: Define priority rules for selecting observation data; for example, for data from source A and source B, it is necessary to configure whether to prioritize using source A alone, source B alone, or the union of data from both sources; this is suitable for comparative experiments that observe different data sources.

[0029] Data quality control strategy: Before entering the analysis, the observation data needs to undergo quality control; the quality control code identifiers of the observation data used in the fusion analysis are parameterized and defined;

[0030] Delay processing strategy: Define a minimum threshold for the number of observations; if the observation delay of an important data source causes the number to fall below the minimum threshold, the process scheduling can make the following decisions: (1) stop waiting and use the existing data for fusion analysis; (2) start a supplementary process that does not depend on the scheduling flow; (3) wait for the number of data to exceed the minimum threshold and then trigger an analysis update.

[0031] The aforementioned workflow-based adaptive scheduling method for surface condition analysis uses the background field as the initial guess field for surface condition analysis, and its selection is crucial to the quality of the surface condition analysis product. Step P6 defines the source and selection strategy of the background field to achieve adaptive use of the background field. Its parameterized definition includes:

[0032] Background field type: Different identifiers are planned for background fields of different sources or resolutions to distinguish them, and these are defined parametrically;

[0033] Background field selection strategy: (1) Based on availability: When a high-resolution background field is missing, the system can automatically downgrade to use a low-resolution background field as a substitute to ensure the timely output of analysis products and reduce the dependence on a single background field; (2) Based on timeliness: Based on the timeliness of the background field, the available background field is selected first; when a higher quality background field is ready, a fusion analysis scheduling based on a high-quality background field can be triggered again.

[0034] Background field preprocessing strategy: The selected background field may need to be preprocessed (such as interpolation, region clipping, etc.) before fusion analysis. Therefore, the preprocessing algorithm itself can also be used as a configurable parameter and selected according to the requirements of the fusion analysis product.

[0035] By defining the background field of fusion analysis as an optional parameterized resource, the system can balance the quality and timeliness of fusion analysis products according to specific analysis scenarios, data conditions, and processing needs.

[0036] The aforementioned workflow-based adaptive scheduling method for surface condition analysis uses an analysis algorithm as its core engine, fusing observation data with the background field to generate the final surface condition analysis product. Step P7 focuses on "modularizing" and "strategizing" the algorithm, enabling it to be invoked by the adaptive scheduling system. The algorithm definition in step P7 includes the following:

[0037] Algorithm type: Different identifiers are designed to distinguish analysis algorithms for different technical solutions, and parameterization is performed;

[0038] Algorithm selection parameterization: The system scheduling engine selects the configured analysis algorithm for the corresponding workflow instance based on the system adaptive identifier content. This can be configured according to the business scenario objectives or the data conditions of the fusion analysis.

[0039] Adaptive configuration of algorithm parameters: The fusion analysis algorithm itself has adjustable parameters (such as influence radius, weight function, etc.); these key algorithm parameters are parameterized so that the scheduling engine can adaptively configure according to the definition, realize refined analysis, and be suitable for sensitivity test scheduling analysis of different technical solutions in the product development stage.

[0040] An adaptive scheduling device for surface condition analysis based on workflow includes an adaptive scheduling framework definition module, a system adaptive identifier design module, a system operating environment parameterization module, an analysis time parameterization module, an analysis observation data definition module, an analysis background field definition module, and an analysis algorithm definition module.

[0041] The adaptive scheduling framework definition module implements step P1 in the above-mentioned workflow-based adaptive scheduling method for surface condition analysis; the system adaptive identifier design module implements step P2 in the above-mentioned workflow-based adaptive scheduling method for surface condition analysis; the system operating environment parameterization module implements step P3 in the above-mentioned workflow-based adaptive scheduling method for surface condition analysis; the analysis time parameterization module implements step P4 in the above-mentioned workflow-based adaptive scheduling method for surface condition analysis; the analysis observation data definition module implements step P5 in the above-mentioned workflow-based adaptive scheduling method for surface condition analysis; the analysis background field definition module implements step P6 in the above-mentioned workflow-based adaptive scheduling method for surface condition analysis; and the analysis algorithm definition module implements step P7 in the above-mentioned workflow-based adaptive scheduling method for surface condition analysis.

[0042] An application of a workflow-based adaptive scheduling method for surface condition analysis is presented, which is used in business scenarios such as surface condition analysis product development, case evaluation, business system construction, and historical product backtesting.

[0043] The technical solution of the present invention achieves the following beneficial technical effects:

[0044] 1. This invention relates to an adaptive scheduling method for surface condition analysis based on workflow. Aimed at the development and system construction of surface condition analysis products, it abstracts surface condition analysis tasks such as observation data preprocessing, background field processing, and fusion analysis into programmable workflow nodes. System identifiers are embedded between these nodes for adaptive scheduling. Through parameterization and module reuse technologies, it achieves the flexibility required for scheduling surface condition analysis products at various stages, including R&D, evaluation, sensitivity testing, business operation, and historical product backtesting. Furthermore, the system identifiers allow for the splitting of a global process into multiple independently executable sub-processes, enabling horizontal scaling to shorten product production time without interrupting upstream and downstream business.

[0045] 2. This invention provides a workflow-based adaptive scheduling method for surface condition analysis. This method enables unified scheduling and management of all task modules within the surface condition analysis framework. Furthermore, this adaptive scheduling approach addresses the code redundancy and redundant system construction issues inherent in traditional static scheduling, effectively improving the reusability of task modules and system construction efficiency. The workflow decomposition mechanism allows for horizontal expansion of the workflow, thereby effectively shortening product production time. This workflow-based adaptive scheduling method for surface condition analysis effectively enhances the automation level, resource utilization efficiency, and ability to handle complex scenarios in the condition analysis system. It can be applied to business scenarios such as surface condition analysis product development, case evaluation, business system construction, and historical product backtesting. Attached Figure Description

[0046] Figure 1 An adaptive scheduling flowchart based on surface condition analysis in this embodiment of the invention. Detailed Implementation

[0047] like Figure 1 As shown, the adaptive scheduling method based on surface condition analysis of workflow in this embodiment includes the following steps:

[0048] Step P1: Define the adaptive scheduling framework: Based on the development content of the surface condition analysis product, define the adaptive scheduling framework to form a tree-like hierarchical structure definition framework file corresponding to the surface condition analysis product; this provides the basic structure and framework for subsequent parameterization and adaptive scheduling; specifically, it includes the following:

[0049] (P1-1) Based on the surface condition analysis, define the corresponding suite (work package) name and framework content for the product development content;

[0050] (P1-2) Define the family name and framework content for different scenarios within the suite (job package) node framework. These generally include rapid operation scenario, standard operation scenario, product rework scenario, historical product back-calculation scenario, and different sensitive test evaluation scenarios, etc.

[0051] (P1-3) Within the family (job family) node framework for dealing with different scenarios, the process is planned as observation data preprocessing, background field processing, fusion analysis and product postprocessing, etc., and the corresponding sub-family (sub-job family) content or task (task) elements are defined according to the processing logic.

[0052] (P1-4) Adopting a modular structural design, the functional modules involved in each scenario in the surface reality fusion analysis are extracted to complete the definition of the name and execution function framework of the task module for a specific function;

[0053] (P1-5) Classify the logically combinable task functional modules and use sub-family (sub-job family) for framework definition and representation.

[0054] Step P2, System Adaptive Identifier Design: Based on the established adaptive scheduling framework, plan and design the system adaptive identifier. The system adaptive identifier includes the surface reality analysis system identifier, business scenario identifier, application scheme identifier, and sequence number identifier.

[0055] The surface condition analysis system identifier is used to distinguish different condition analysis systems; the business scenario identifier is used to distinguish different business scenarios such as Rapid Run, Cycle Run, Rework Run, Historical Product Backtesting (HIST), and Sensitive Test (TC); the application solution identifier is used to distinguish the different surface condition analysis technology solutions adopted; the serial number identifier is optional and uses two digits; it is mainly used to identify different operation scheduling branches when the surface condition analysis system identifier, business scenario identifier, and application solution identifier are the same, and when the process is split and parallelized, the surface condition analysis system identifier, business scenario identifier, application solution identifier, and serial number identifier are separated by an underscore "_";

[0056] By parsing the system identifier content, the adaptive scheduling configuration script can accurately match the scheduled running tasks with the corresponding environmental resources, data, background fields, fusion methods, etc. at the appropriate running time, thereby realizing process scheduling and adaptation.

[0057] Step P3, System Operating Environment Parameterization: Parameterize the operating environment of each system adaptive identifier and generate a corresponding operating environment configuration file. The operating environment configuration file defines and declares the parameterized content of the system adaptive identifier's operating environment, including the main directory for system script execution, the directory for observation data input and quality control processing, the directory for background field input and processing, the directory for fusion analysis product generation, and the directory for product post-processing. Depending on the business scenario, each adaptive identifier can use its own independent parameterized configuration definition content or share the same parameterized configuration definition content.

[0058] Step P4, System Analysis Time Parameterization: The analysis time of the surface condition analysis system is parameterized from three aspects: analysis time schedule, basic analysis cycle, and analysis time window; among which,

[0059] The method for customizing the analysis schedule is as follows: the analysis schedule is set using a Crontab-like expression to achieve automatic processing of the product analysis cycle; for example, it can be defined to perform the current time period analysis every 3 hours with a 45-minute lag.

[0060] The method for defining the basic analysis cycle is as follows: set the basic analysis time interval, for example, start it once every 20 minutes, and determine the specific time of surface condition analysis based on product efficiency, observation data, background field data, and analysis of product generation.

[0061] Dynamic adjustment of analysis time windows: By parameterizing the start and end times of analysis time windows, the analysis time windows can be adjusted according to the actual needs of the business scenario, thereby starting the required real-time analysis tasks at the appropriate time and improving the timeliness of the analysis.

[0062] Step P5: Define and classify the analytical observation data of the surface condition analysis system. Observation data is one of the important input data for condition analysis. Refining its definition and classification aims to enable the system to adaptively identify, select, and preprocess observation data from multiple sources. Especially when accessing observation data of the same type but from different sources, targeted data definition and preprocessing configuration are required. This invention defines and classifies analytical observation data from the following dimensions:

[0063] Data source type: Different identifiers are designed for observation data from different data sources to distinguish them, and these identifiers are defined parametrically.

[0064] Prioritization strategy: Define priority rules for selecting observation data; for example, for data from source A and source B, it is necessary to configure whether to prioritize using source A alone, source B alone, or the union of data from both sources; this is suitable for comparative experiments that observe different data sources.

[0065] Data quality control strategy: Before entering the analysis, the observation data needs to undergo quality control; the quality control code identifiers of the observation data used in the fusion analysis are parameterized and defined;

[0066] Delay processing strategy: Define a minimum threshold for the number of observations; if the observation delay of an important data source causes the number to fall below the minimum threshold, the process scheduling can make the following decisions: (1) stop waiting and use the existing data for fusion analysis; (2) start a supplementary process that does not depend on the scheduling flow; (3) wait for the number of data to exceed the minimum threshold and then trigger an analysis update.

[0067] Step P6, Background Field Definition: This step involves parameterizing the background field type and selection strategy for the surface condition analysis system. The background field provides the initial guess field for the surface condition analysis, and its selection is crucial to the quality of the surface condition analysis product. The source and selection strategy of the background field are defined to achieve adaptive use. The parameterized definition of the background field in this implementation includes:

[0068] Background field selection strategy: (1) Based on availability: When a high-resolution background field is missing, the system can automatically downgrade to use a low-resolution background field as a substitute to ensure the timely output of analysis products and reduce the dependence on a single background field; (2) Based on timeliness: Based on the timeliness of the background field, the available background field is selected first; when a higher quality background field is ready, a fusion analysis scheduling based on a high-quality background field can be triggered again.

[0069] Background field preprocessing strategy: The selected background field may need to be preprocessed (such as interpolation, region clipping, etc.) before fusion analysis. Therefore, the preprocessing algorithm itself can also be used as a configurable parameter and selected according to the requirements of the fusion analysis product.

[0070] By defining the background field of fusion analysis as an optional parameterized resource, the system can balance the quality and timeliness of fusion analysis products according to specific analysis scenarios, data conditions, and processing needs.

[0071] Step P7, Algorithm Definition: The analysis algorithm of the surface condition analysis system is parameterized to achieve modularization and strategic application. The analysis algorithm is the core engine for fusing observation data with the background field to generate the final surface condition analysis product. The core of defining the analysis algorithm is to "modularize" and "strategize" the algorithm, enabling it to be invoked by the adaptive scheduling system. In this embodiment, the algorithm definition includes the following:

[0072] Algorithm type: Different identifiers are designed to distinguish analysis algorithms for different technical solutions, and parameterization is performed;

[0073] Algorithm selection parameterization: The system scheduling engine selects the configured analysis algorithm for the corresponding workflow instance based on the system adaptive identifier content. This can be configured according to the business scenario objectives or the data conditions of the fusion analysis.

[0074] Adaptive configuration of algorithm parameters: The fusion analysis algorithm itself has adjustable parameters (such as influence radius, weight function, etc.); these key algorithm parameters are parameterized so that the scheduling engine can adaptively configure according to the definition, realize refined analysis, and be suitable for sensitivity test scheduling analysis of different technical solutions in the product development stage.

[0075] Applying the workflow-based adaptive scheduling method for surface condition analysis in this embodiment to the research and development of global 10km hourly surface condition analysis products and atmospheric chemical surface condition analysis products in China, as well as the construction of business systems, can significantly improve the automation level and resource utilization efficiency of each business product, shorten product production time, and improve system construction efficiency.

[0076] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of the claims of this patent application.

Claims

1. An adaptive scheduling method based on workflow-based surface condition analysis, characterized in that, Includes the following steps: Step P1: Define an adaptive scheduling framework based on the development content of the surface condition analysis product, and form a tree-structured definition content file corresponding to the surface condition analysis product; this provides the basic structure and framework for subsequent parameterization and adaptive scheduling. Step P2: Based on the established adaptive scheduling framework, plan and design the system adaptive identifiers. The system adaptive identifiers include the surface reality analysis system identifier, business scenario identifier, application scheme identifier, and sequence number identifier. Step P3: Parameterize the runtime environment of each system adaptive identifier and generate the corresponding runtime environment configuration file; Step P4: Parameterize the analysis time of the surface condition analysis system from three aspects: analysis schedule, basic analysis cycle, and analysis time window; Step P5: Define and classify the analytical observation data of the surface condition analysis system; Step P6: Parameterize the background field type and background field selection strategy of the surface condition analysis system; Step P7: Parameterize the analysis algorithm of the surface condition analysis system to achieve modularization and strategy-based analysis algorithm.

2. The adaptive scheduling method based on workflow-based surface condition analysis according to claim 1, characterized in that, The definition of the adaptive scheduling framework in step P1 includes the following: (P1-1) Based on the surface condition analysis, define the corresponding work package name and framework content for the product development content; (P1-2) Within the work package node framework, define the work family names and framework content for different scenarios, including rapid operation scenario, standard operation scenario, product rework scenario, historical product back-calculation scenario, and different sensitive test evaluation scenarios. (P1-3) Within the framework of job family nodes for dealing with different scenarios, the process is planned as observation data preprocessing, background field processing, fusion analysis and product post-processing, and the corresponding sub-job family content or task elements are defined according to the processing logic. (P1-4) Adopting a modular structural design, the functional modules involved in each scenario in the surface reality fusion analysis are extracted to complete the definition of the task functional module name and execution functional framework; (P1-5) Classify the logically combinable task function modules and use sub-job families for framework definition and representation.

3. The adaptive scheduling method based on workflow-based surface condition analysis according to claim 1, characterized in that, In step P2, the surface condition analysis system identifier is used to distinguish different condition analysis systems; the business scenario identifier is used to distinguish between rapid operation scenarios, standard operation scenarios, product rework scenarios, historical product backtesting scenarios, and different sensitive test evaluation scenarios; the application solution identifier is used to distinguish the content of different surface condition analysis technical solutions; the serial number identifier uses two digits to identify different operation scheduling branches when the surface condition analysis system identifier, business scenario identifier, and application solution identifier are in the same context for process splitting and parallel processing; the surface condition analysis system identifier, business scenario identifier, application solution identifier, and serial number identifier are separated by an underscore symbol "_".

4. The adaptive scheduling method for surface condition analysis based on workflow according to claim 1, characterized in that, In step P3, the runtime environment configuration file defines and declares the parameterized content of the system adaptive identification runtime environment, including the main directory for system script execution, the directory for observation data input and quality control processing, the directory for background field input and processing, the directory for fusion analysis product generation, and the directory for product post-processing.

5. The adaptive scheduling method for surface condition analysis based on workflow according to claim 1, characterized in that, In step P4, the method for customizing the analysis schedule is as follows: the analysis schedule is set by defining the content of the Crontab-like expression to achieve automatic processing of the product analysis cycle; The method for defining the basic analysis cycle is as follows: set the basic analysis time interval, and determine the specific time of surface condition analysis based on product efficiency, observation data, background field data, and analysis of product generation. Dynamic adjustment of analysis time windows: The start and end times of analysis time windows can be parameterized and dynamically adjusted according to the actual needs of the business scenario.

6. The adaptive scheduling method for surface condition analysis based on workflow according to claim 1, characterized in that, In step P5, the analysis of the observation data definition and classification is carried out from the following dimensions: Data source type: Different identifiers are designed for observation data from different data sources to distinguish them, and these identifiers are defined parametrically; Priority strategy: Priority rules for selecting observation data are established; Data quality control strategy: Quality control code identifiers for observation data used in fusion analysis are configured and defined parametrically. Delay processing strategy: Define a minimum threshold for the number of observations; If the data source observation delay causes the number to fall below the minimum threshold, the process scheduling will make the following decisions: (1) stop waiting and use the existing data for fusion analysis; (2) start the supplementary process; (3) wait for the number of observations in the analysis observation data to exceed the minimum threshold before triggering an analysis update.

7. The adaptive scheduling method for surface condition analysis based on workflow according to claim 1, characterized in that, In step P6, the parameterization definition includes: Background field type: Different identifiers are planned for background fields of different sources or resolutions to distinguish them, and these are defined parametrically; Background field selection strategy: (1) Based on availability: When a high-resolution background field is missing, the system automatically downgrades to use a low-resolution background field as a substitute; (2) Based on timeliness: Based on the timeliness of the background field, the available background field is selected first. When a higher quality background field is ready, a fusion analysis scheduling based on a high-quality background field is triggered again. Background field preprocessing strategy: The preprocessing algorithm itself is treated as a configurable parameter and selected according to the requirements of the fusion analysis product.

8. The adaptive scheduling method for surface condition analysis based on workflow according to claim 1, characterized in that, The algorithm definition in step P7 includes the following: Algorithm type: Different identifiers are designed to distinguish analysis algorithms for different technical solutions, and parameterization is performed; Algorithm selection parameterization: The system scheduling engine selects the configured analysis algorithm for the corresponding workflow instance based on the system adaptive identifier content, and configures it according to the business scenario objectives or the data conditions of the fusion analysis; Adaptive configuration of algorithm parameters: The fusion analysis algorithm itself has adjustable parameters, including the influence radius and weight function.

9. A workflow-based adaptive scheduling device for surface condition analysis, characterized in that, It includes an adaptive scheduling framework definition module, a system adaptive identifier design module, a system operating environment parameterization module, an analysis time parameterization module, an analysis observation data definition module, an analysis background field definition module, and an analysis algorithm definition module; The adaptive scheduling framework definition module is used to implement step P1 in the workflow-based adaptive scheduling method for surface condition analysis as described in claim 1. The system adaptive identifier design module is used to implement step P2 in the workflow-based adaptive scheduling method for surface condition analysis as described in claim 1. The system runtime environment parameterization module is used to implement step P3 in the workflow-based adaptive scheduling method for surface condition analysis as described in claim 1. The analysis time parameterization module is used to implement step P4 in the workflow-based adaptive scheduling method for surface condition analysis as described in claim 1. The analysis and observation data definition module is used to implement step P5 in the workflow-based adaptive scheduling method for surface condition analysis as described in claim 1. The background field definition module is used to implement step P6 in the workflow-based adaptive scheduling method for surface condition analysis as described in claim 1. The analysis algorithm definition module is used to implement step P7 in the workflow-based adaptive scheduling method for surface condition analysis as described in claim 1.

10. An application of an adaptive scheduling method based on workflow-based surface condition analysis, characterized in that, The workflow-based adaptive scheduling method for surface condition analysis as described in any one of claims 1-8 can be applied to business scenarios such as surface condition analysis product development, case evaluation, business system construction, and historical product backtesting.

Images