A low-code software automation task orchestration engine system and implementation method

Abstract

The application discloses a kind of low code software automation task orchestration engine systems and implementation methods, it is related to task orchestration engine technical field, comprising: atomic component library module provides pre-packaged dragable atomic component, support workflow construction;Workflow orchestration module constructs workflow, completes three-state life cycle and version management, and exports workflow metadata;Multi-source scheduling trigger module adapts five kinds of trigger mode through unified scheduling model, generates task instance and issues execution instruction when trigger condition is satisfied;Execution monitoring module drives workflow node sequential execution, collects full-link log, real-time updates task instance state and returns execution data;Workbench board module completes index statistics, trend visualization, high-frequency workflow and abnormal risk screening;The low code software automation task orchestration engine system and implementation method provided by the application solve the problems of high development threshold, fragmented trigger mode and poor process reuse in the prior art.

Description

Technical Field

[0001] This invention relates to the field of task orchestration engine technology, and in particular to a low-code software automated task orchestration engine system and its implementation method. Background Technology

[0002] As enterprises continue to advance digitalization, they typically deploy multiple heterogeneous business systems, generating numerous demands for data synchronization, event response, and business collaboration between these systems. The forms of automated business processes have also expanded from traditional scheduled batch processing to diverse scenarios such as interface callbacks, event-driven processes, and human-machine collaborative review. How to build, schedule, execute, and maintain automated processes at low cost has become a pressing technical challenge in enterprise digital transformation.

[0003] Currently, the main solutions for business automation in the industry include native code development, general task scheduling frameworks, commercial business process management (BPM) platforms, and low-code workflow tools. While these solutions can achieve basic automation capabilities, they all have inherent shortcomings to varying degrees due to limitations in their underlying architecture, making it difficult to simultaneously meet the comprehensive business needs in complex scenarios.

[0004] Native code development solutions offer high flexibility but heavily rely on specialized developers. Business personnel cannot maintain processes independently, business change iteration cycles are long, code reusability is poor, troubleshooting is difficult, and the cost of implementing large-scale automated business processes remains high. General task scheduling frameworks focus on scheduled task execution, with a single triggering mode. They lack native normalization adaptation for Webhooks and external business event-driven processes, and the scheduling, execution, and log statistics modules are fragmented, lacking a standardized data feedback loop. This prevents the formation of a closed-loop management system encompassing orchestration, scheduling, execution, and monitoring, and operations and maintenance rely on manually exporting multi-source data for aggregation and analysis.

[0005] Traditional BPM platforms, centered on manual workflow approvals, are ill-suited for automated scheduling between systems. While existing low-code workflow tools offer visual orchestration, they generally lack standardized, layered atomic components and robust sub-workflow reuse mechanisms. Modifications to common business logic require manual synchronization in multiple locations. Observability is limited to post-event log queries for single tasks, lacking global metric statistics and the ability to predict risks in high-frequency processes. Furthermore, the automated execution chain is disconnected from human-machine approval capabilities, making it difficult to simultaneously meet the dual needs of automated system execution and business compliance review. Summary of the Invention

[0006] The purpose of this invention is to provide a low-code software automation task orchestration engine system and implementation method to solve the above-mentioned problems existing in the prior art.

[0007] To achieve the above objectives, the present invention provides a low-code software automation task orchestration engine system, including a workbench Kanban module, a workflow orchestration module, an atomic component library module, an execution monitoring module, and a multi-source scheduling triggering module; The atomic component library module stores and provides standardized, draggable atomic components to support the workflow orchestration module, which can then call these components to build a visual representation of automated workflows. The workflow orchestration module is the core orchestration carrier of the system. It is used to respond to the user's visual drag-and-drop operation, call various atomic components provided by the atomic component library module to build and store DAG structured automated workflows, implement three-state lifecycle management and iteration version management for workflows (draft, published, and deactivated), and synchronously output the metadata of the published workflows that have passed the verification to the multi-source scheduling trigger module to complete the pre-distribution of process resources. The multi-source scheduling trigger module establishes a data communication connection with the workflow orchestration module to receive and bind metadata of published workflows. It adapts to five triggering methods—timed triggering, fixed-interval triggering, Webhook interface triggering, business event-driven triggering, and manual triggering—through a unified scheduling model. It monitors the triggering conditions of various triggering methods in real time. When the triggering conditions are met, it outputs workflow execution instructions to the execution monitoring module and initializes and creates corresponding task instances. The execution monitoring module is connected to the multi-source scheduling trigger module, the workflow orchestration module, and the workbench dashboard module. It is used to receive execution instructions issued by the multi-source scheduling trigger module, load the workflow structure data stored in the workflow orchestration module, drive each node of the workflow to execute sequentially, collect and record the input parameters, output results and full-link operation logs of each node throughout the process, update the running status of task instances in real time, and aggregate all task instance running data and transmit it back to the workbench dashboard module in real time. The workbench dashboard module is used to receive and aggregate all task instance running data pushed by the execution monitoring module, and generate statistical core workflow indicator data, periodic execution trend data and high-frequency workflow ranking data. It realizes the visualization of the health status of automated task instances across the entire system, and achieves closed-loop linkage management and control of the entire lifecycle of process orchestration, scheduling triggering, task instance running, monitoring and observation, and process iteration.

[0008] This invention also provides a method for implementing a low-code software automation task orchestration engine system, comprising the following steps: S1. The workflow orchestration module responds to the user's visual drag-and-drop operation, calls various atomic components of the atomic component library module to build and generate a DAG-structured automated workflow, completes the parameter configuration and logic verification of each node, publishes and generates the corresponding iteration version of the workflow, and synchronizes the workflow's metadata to the multi-source scheduling trigger module. S2. The multi-source scheduling trigger module receives the metadata of the published workflow and completes the binding association, configures single or multiple trigger conditions for the workflow, turns on the scheduling activation switch, and completes the initial configuration of the trigger method. S3: The multi-source scheduling trigger module polls and monitors various triggering methods in real time. When any of the bound triggering conditions are met, it generates the corresponding task instance and execution instruction, and pushes it to the execution monitoring module to wait for execution. S4. The execution monitoring module receives execution instructions and task instance information, loads the structure and parameter configuration of the corresponding workflow, drives each node of the workflow to execute sequentially, records the entire chain operation log and node interaction data, and updates the task instance running status in real time. S5, the execution monitoring module aggregates all task instance running data and pushes it to the workbench dashboard module in real time. The workbench dashboard module completes data summarization, indicator statistics, trend visualization and health analysis, realizing closed-loop management of the entire lifecycle of automated task instances from orchestration, scheduling, execution to monitoring iteration.

[0009] Therefore, the present invention employs the above-mentioned low-code software automation task orchestration engine system and implementation method, which has the following beneficial effects: (1) Adopting a low-code architecture that combines pre-packaged atomic components with visual drag-and-drop, business personnel can build DAG structured automated workflows mainly by configuring parameters, and only retain script execution nodes for local logic expansion, thereby reducing the threshold for building automated processes, reducing reliance on professional coders, and improving business iteration efficiency; (2) Construct a unified scheduling model that normalizes multiple trigger sources, resolve conflicts and encapsulate five types of trigger signals: timed, fixed interval, Webhook interface, business event driven, and manual. Support binding multiple heterogeneous trigger rules to a single workflow. Task instances generated by various triggers are integrated into the same execution and statistics system. Avoid the problems of fragmented trigger modes and difficulty in unified management of multi-source events in traditional solutions. Adapt to diverse cross-system linkage scenarios of enterprises. (3) Relying on the high-frequency workflow ranking unit, global risk proactive screening and early warning are achieved in two dimensions: execution frequency and success rate; the execution monitoring module saves the full-link log of task instances, which can accurately locate fault nodes, upgrade the traditional post-event passive investigation to global risk prediction, and improve the efficiency of operation and maintenance. (4) Achieve deep integration of automated processes and human-machine approval. Manual approval nodes can be flexibly embedded in any position of the workflow to complete the pause pending review, review resume or terminate the process. Approval data is fully collected and participates in global statistics, taking into account both the system's automatic execution efficiency and the enterprise's business compliance review needs. (5) The five modules form a closed-loop linkage of data, realizing a complete link of process orchestration, scheduling triggering, task instance execution, monitoring and observation and process iteration. Data between modules automatically flows back and aggregates, eliminating the need for manual export of multi-source data and manual statistics, and truly realizing automated closed-loop management of the entire business lifecycle.

[0010] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the structure of a low-code software automation task orchestration engine system according to the present invention; Figure 2 This is an overall flowchart of the implementation method of a low-code software automation task orchestration engine system according to the present invention. Detailed Implementation

[0012] The following detailed description of embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0013] Please see Figure 1 A low-code software automation task orchestration engine system includes a workbench Kanban module, a workflow orchestration module, an atomic component library module, an execution monitoring module, and a multi-source scheduling trigger module.

[0014] This system implements a low-code working mode: various business execution logics are pre-encapsulated within atomic components, and users build DAG structured automated workflows mainly through visual drag-and-drop and parameter configuration, without having to write complete business code by hand; it only opens up local custom code extension capabilities through script execution nodes, realizing a low-code orchestration mode that prioritizes configuration and supplements with a small amount of code.

[0015] The Atomic Component Library module stores and provides standardized, draggable atomic components of various types, providing underlying component support for the workflow orchestration module, which can then call these components to complete the visual construction of automated workflows.

[0016] The atomic component library module contains four types of layered atomic components: process control components, integration and connection components, data processing components, and human collaboration components. Process control components manage the overall execution logic of the workflow, including start nodes, end nodes, conditional branch nodes, delay / wait nodes, loop traversal nodes, and sub-workflow call nodes. Integration and connection components enable cross-platform and cross-device data interaction, including HTTP request nodes, message notification nodes, and database operation nodes. Data processing components clean, format, and transform various types of raw business data, including script execution nodes and data transformation nodes. Human collaboration components are human approval nodes, used to implement the human-machine collaborative execution logic of pausing, awaiting review, manually reviewing, and resuming approved workflows. Sub-workflow call nodes are used to call published workflows in the workflow orchestration module, enabling one-time orchestration of common business logic and reuse across multiple scenarios. Furthermore, after a workflow undergoes version iteration updates, the process logic of all related call scenarios is automatically updated synchronously, eliminating the need for repeated manual orchestration and configuration. The manual approval node can be flexibly configured between any execution node of the automated workflow. When the workflow reaches the manual approval node, the execution link is automatically paused and an approval notification is pushed. After receiving the instruction of manual approval or rejection, the workflow operation is adaptively resumed or terminated. All approval process data, approval results, and workflow operation data are synchronously stored in the execution monitoring module and uniformly included in the task instance success rate statistics of the workbench dashboard module.

[0017] The workflow orchestration module is the core orchestration carrier of the system. It is used to respond to the user's visual drag-and-drop operation, call various atomic components provided by the atomic component library module to build and store DAG structured automated workflows, implement three-state lifecycle management and iteration version management for workflows (draft, published, and deactivated), and synchronously output the metadata of the published workflows that have passed the verification to the multi-source scheduling trigger module to complete the pre-deployment of process resources.

[0018] The multi-source scheduling trigger module establishes a data communication connection with the workflow orchestration module to receive and bind metadata of published workflows. It adapts to five triggering methods—timed triggering, fixed-interval triggering, Webhook interface triggering, business event-driven triggering, and manual triggering—through a unified scheduling model. It monitors the triggering conditions for each method in real time. When the triggering conditions are met, it outputs workflow execution instructions to the execution monitoring module and initializes and creates corresponding task instances. Based on a unified scheduling data model, the multi-source scheduling trigger module performs unified integrated management of the five triggering methods. It supports binding multiple different types of triggering conditions to a single published workflow simultaneously. All triggering conditions are centrally managed for start / stop. Task instances generated by various triggering methods are uniformly imported into the execution monitoring module for log collection, status updates, and data statistics. The specific triggering conditions for the five triggering methods are as follows: Timed triggering condition: the system time matches a preset Cron time expression; Fixed interval triggering condition: the system runtime reaches a preset fixed time period; Webhook interface triggering condition: receiving a valid HTTP callback request pushed by an external system; Business event-driven triggering condition: detecting a preset subscribed business event identifier and event message; Manual triggering condition: receiving a process execution instruction actively issued by the front-end user. The expression for the unified scheduling data model is: ; In the formula, To normalize and unify the triggering timing of the workflow; It is a multi-trigger source normalization fusion scheduling operator used to resolve conflicts, adapt priorities, and encapsulate the state of five independent trigger signals, so as to realize the parallel management and unified output of multiple types of trigger rules; For timed triggering; For fixed interval triggering timing; The timing for triggering the Webhook interface; For business event-driven triggering timing; This is for manual triggering. The multi-trigger source normalization fusion scheduling operator... The processing logic is as follows: receive independent trigger signals from five types of trigger sources, perform signal deduplication, resolve conflicts between multiple trigger conditions, adapt trigger source priorities, encapsulate the state of trigger signals from different sources, and output a normalized unified trigger timing. When multiple different types of triggering conditions for the same automated workflow are met simultaneously, the system can be configured to generate multiple independent task instances in parallel, or to enable a debouncing and deduplication strategy to avoid repeated triggering in a short period of time.

[0019] The execution monitoring module communicates with the multi-source scheduling trigger module, workflow orchestration module, and workbench dashboard module. It receives execution commands from the multi-source scheduling trigger module, loads workflow structure data stored in the workflow orchestration module, drives the sequential execution of each workflow node, collects and records input parameters, output results, and end-to-end logs for each node, updates the task instance running status in real time, and aggregates all task instance running data and transmits it back to the workbench dashboard module in real time. The execution monitoring module assigns a unique execution ID to each workflow execution behavior, corresponding to the storage of the task instance's triggering method, triggering entity, start time, running time, execution status, running summary, and detailed logs for all nodes. It supports forced termination of running task instances and can accurately locate process failure nodes and anomalies based on the end-to-end running logs.

[0020] The workbench dashboard module is used to receive and aggregate all task instance running data pushed by the execution monitoring module, and generate statistical core workflow indicator data, periodic execution trend data and high-frequency workflow ranking data. It realizes the visualization of the health status of automated task instances across the entire system, and achieves closed-loop linkage management and control of the entire lifecycle of process orchestration, scheduling triggering, task instance running, monitoring and observation, and process iteration.

[0021] The workbench dashboard module includes a metrics statistics unit, a periodic trend visualization unit, a high-frequency workflow ranking unit, and a recently executed instance unit. The metrics statistics unit displays the total number of workflows, the number of published workflows, the total number of task instance executions for the day, the task instance success rate, the number of active scheduling rules, and the number of running task instances. The periodic trend visualization unit generates a visualization chart showing the execution trend of task instances over the past seven days, indicating success and failure. The high-frequency workflow ranking unit sorts workflows in descending order of cumulative task instance executions, displays high-frequency workflows and their corresponding task instance success rates, and quickly screens for global abnormal and risky processes. Specifically, it includes marking, locating, and issuing risk warnings for high-frequency, low-success-rate abnormal workflows, accurately identifying frequent errors and unstable operations. The core automated workflow for business adaptation anomalies enables proactive prediction and rapid location of global business risks. High-frequency workflows are defined as published automated workflows that, within a preset statistical period, have a cumulative execution count exceeding a system-preset frequency threshold and are in a high-frequency business call state; these are the process resources corresponding to the enterprise's core high-frequency automated business. The most recently executed instance unit displays multiple newly generated task instance records in real time, intuitively presenting the execution time, associated workflow, running status, and execution summary of each task instance, allowing operations personnel to quickly view recent automated execution details and locate sudden abnormal task instances.

[0022] This system internally maintains three types of core logical data objects: (1) DAG structured automated workflow object: includes a node set and an edge set; the node set stores nodeId, component type, and node configuration parameters; the edge set stores the starting node identifier and the target node identifier, which are used to describe the execution order dependency between nodes; (2) Task instance object: includes a globally unique execution ID, the workflow ID to which it belongs, the trigger method identifier, and an enumeration of task instance status. The task instance status includes pending execution, in execution, awaiting approval, success, failure, and termination; and also saves the input and output logs of each node. (3) Scheduling configuration object: includes the bound workflow identifier, trigger type, corresponding trigger condition configuration parameters, and enable switch flag.

[0023] The preset statistical period for high-frequency workflows is 7 days, and the preset frequency threshold is 20 times for example. If the cumulative number of executions of published workflow task instances is not less than 20 within the 7-day period, it is determined to be a high-frequency workflow. The success rate alarm threshold for abnormal risk screening is 85% for example. When the success rate of the task instance corresponding to a high-frequency workflow is lower than 85%, the high-frequency workflow ranking unit marks the workflow with an early warning. The above frequency threshold and success rate alarm threshold can be customized by operation and maintenance personnel in the system interface.

[0024] This invention also covers abnormal execution scenarios: when a component node within a workflow fails to execute, the execution monitoring module records the abnormal information of the corresponding node, updates the current task instance status to failure, and completely saves the input, output, and error information of the abnormal node in the full node detailed log; the data of the abnormal task instance is also sent back to the workbench dashboard module to participate in the global success rate statistics; operations and maintenance personnel can identify workflows with abnormal success rates through the workbench, drill down to the execution monitoring module, and locate the specific component node where the error occurred.

[0025] Compared with traditional workflow scheduling systems, this invention does not simply pile up functions such as "drag and drop orchestration, timed scheduling, and log querying". Instead, it forms a complete data feedback link through the closed-loop linkage of five major modules, which integrates component orchestration, multi-source normalized scheduling, task instance execution, global dashboard, and single instance fault location. In contrast, traditional solutions have isolated modules, and the execution results cannot be automatically fed back to the dashboard for risk screening, requiring manual data export for statistics.

[0026] Please see Figure 2 An implementation method for a low-code software automation task orchestration engine system includes the following steps: S1. Workflow Orchestration Setup and Deployment: The workflow orchestration module responds to the user's visual drag-and-drop operation, calls various atomic components from the atomic component library module to build and generate a DAG-structured automated workflow, completes the parameter configuration and logic verification of each node, and publishes the workflow to generate the corresponding iteration version, and synchronizes the workflow's metadata to the multi-source scheduling trigger module. S2. Multi-source scheduling trigger binding and configuration: The multi-source scheduling trigger module receives the metadata of the published workflow and completes the binding association, configures single or multiple trigger conditions for the workflow, turns on the scheduling activation switch, and completes the initial configuration of the trigger method; S3. Polling trigger conditions and generating task instances: The multi-source scheduling trigger module polls and monitors various triggering methods in real time. When any of the bound triggering conditions are met, it generates the corresponding task instance and execution instructions, and pushes them to the execution monitoring module to wait for execution. S4. Drive node execution and log recording: The execution monitoring module receives execution instructions and task instance information, loads the structure and parameter configuration of the corresponding workflow, drives each node of the workflow to execute sequentially, records the entire chain operation log and node interaction data, and updates the task instance running status in real time. S5. Data Aggregation and Dashboard Display: The execution monitoring module aggregates all task instance running data and pushes it to the workbench dashboard module in real time. The workbench dashboard module completes data summarization, indicator statistics, trend visualization, and health analysis, realizing closed-loop management of the entire lifecycle of automated task instances from orchestration, scheduling, execution to monitoring iteration.

[0027] Therefore, this invention adopts the aforementioned low-code software automation task orchestration engine system and implementation method. It employs a low-code architecture combining pre-encapsulated atomic components with visual drag-and-drop functionality. Business personnel can build DAG-structured automated workflows primarily through parameter configuration, retaining only script execution nodes for local logic expansion. This lowers the barrier to building automated processes, reduces reliance on professional coders, and improves business iteration efficiency. A unified scheduling model with multiple trigger sources is constructed, resolving conflicts and unifying the encapsulation of five types of trigger signals: timed, fixed-interval, Webhook interfaces, business event-driven, and manual. It supports binding multiple heterogeneous trigger rules to a single workflow, and task instances generated by various triggers are integrated into the same execution and statistics system. This avoids the fragmentation of trigger modes and the difficulty in unified management of multi-source events in traditional solutions, adapting to diverse cross-system requirements of enterprises. Linked scenarios; relying on high-frequency workflow ranking units, global risk proactive screening and early warning are achieved through execution frequency and success rate; the execution monitoring module saves the full-link log of task instances, which can accurately locate fault nodes, upgrading the traditional passive post-event investigation to global risk prediction and improving operation and maintenance efficiency; it realizes deep integration of automated processes and human-machine approval, and human approval nodes can be flexibly embedded in any position of the workflow to complete the pause pending review, review resume or process termination. Approval data is fully collected and participates in global statistics, taking into account both the system's automatic execution efficiency and the enterprise's business compliance review needs; the five major modules form a data closed loop linkage, realizing a complete link of process orchestration - scheduling trigger - task instance execution - monitoring observation - process iteration. Data between modules automatically flows back and aggregates, eliminating the need for manual export of multi-source data and manual statistics, truly realizing automated closed-loop management of the entire business lifecycle.

[0028] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A low-code software automation task orchestration engine system, characterized in that, include: Workbench Kanban module, workflow orchestration module, atomic component library module, execution monitoring module, multi-source scheduling trigger module; The atomic component library module stores and provides various atomic components, providing underlying component support for the workflow orchestration module, which can then call these components to complete the workflow setup. The workflow orchestration module is used to respond to the user's visual drag-and-drop operation, call various atomic components provided by the atomic component library module to build and store DAG structured workflows, implement three-state lifecycle management and iteration version management for workflows (draft, published, and deactivated), and synchronously output the metadata of the published workflows that have passed the verification to the multi-source scheduling trigger module to complete the pre-distribution of process resources. The multi-source scheduling trigger module establishes a data communication connection with the workflow orchestration module to receive and bind metadata of published workflows. It adapts to five triggering methods—timed triggering, fixed-interval triggering, Webhook interface triggering, business event-driven triggering, and manual triggering—through a unified scheduling model. It monitors the triggering conditions of various triggering methods in real time. When the triggering conditions are met, it outputs workflow execution instructions to the execution monitoring module and initializes and creates corresponding task instances. The execution monitoring module is connected to the multi-source scheduling trigger module, the workflow orchestration module, and the workbench dashboard module. It is used to receive execution instructions issued by the multi-source scheduling trigger module, load the workflow structure data stored in the workflow orchestration module, drive each node of the workflow to execute sequentially, collect and record the input parameters, output results and full-link operation logs of each node throughout the process, update the running status of task instances in real time, and aggregate all task instance running data and transmit it back to the workbench dashboard module in real time. The workbench dashboard module is used to receive and aggregate task instance running data pushed by the execution monitoring module, and generate statistical data on core workflow indicators, periodic execution trend data, and high-frequency workflow ranking data, so as to realize the visual display of the health status of automated task instances across the entire system.

2. The low-code software automation task orchestration engine system according to claim 1, characterized in that: The atomic component library module contains four types of layered atomic components: process control components, integration and connection components, data processing components, and human collaboration components. The process control component is used to manage the overall execution logic of the workflow, including start nodes, end nodes, conditional branch nodes, delay waiting nodes, loop traversal nodes, and sub-workflow call nodes. Integrated connectivity components are used to enable cross-platform and cross-device data interaction and connection of the system, including HTTP request nodes, message notification nodes, and database operation nodes; The data processing component is used to clean, format, and transform various types of raw business data, including script execution nodes and data transformation nodes; The human collaboration component is a human approval node used to implement the human-machine interaction execution logic of workflow pause for review, human review, and review resumption.

3. The low-code software automation task orchestration engine system according to claim 2, characterized in that: Sub-workflow call nodes are used to call published workflows in the workflow orchestration module, enabling one-time orchestration of common business logic and reuse across multiple scenarios. After the workflow completes version iteration updates, the process logic of all related call scenarios is automatically updated synchronously.

4. The low-code software automation task orchestration engine system according to claim 3, characterized in that: The multi-source scheduling trigger module integrates and manages five triggering methods based on a unified scheduling data model. It supports binding multiple different types of triggering conditions to a single published workflow. All triggering conditions are centrally managed for start and stop. Task instances generated by various triggering methods are uniformly imported into the execution monitoring module.

5. A low-code software automation task orchestration engine system according to claim 4, characterized in that, The expression for the unified scheduling data model is: ； In the formula, To normalize and unify the triggering timing of the workflow; For multi-trigger source normalized fusion scheduling operators; For timed triggering; For fixed interval triggering timing; The timing for triggering the Webhook interface; For business event-driven triggering timing; This is a manually triggered event.

6. A low-code software automation task orchestration engine system according to claim 5, characterized in that, The specific triggering conditions for the five triggering methods are as follows: the timed triggering condition is that the system time matches the preset Cron time expression; the fixed interval triggering condition is that the system runtime reaches the preset fixed time period; the Webhook interface triggering condition is that a valid HTTP callback request pushed by an external system is received; and the business event-driven triggering condition is that a preset subscribed business event identifier and event message are detected. The manual trigger condition is receiving a process execution instruction actively issued by the front-end user.

7. A low-code software automation task orchestration engine system according to claim 6, characterized in that: The execution monitoring module assigns a unique execution ID to each workflow operation, corresponding to the triggering method, triggering subject, start time, running time, execution status, running summary, and full node log of the stored task instance. It supports the forced termination of running task instances and locates process fault nodes and abnormal causes based on the full-link running log.

8. A low-code software automation task orchestration engine system according to claim 7, characterized in that: The workbench dashboard module includes a metric statistics unit, a periodic trend visualization unit, a high-frequency workflow ranking unit, and a recently executed instance unit; The metrics statistics unit is used to count and display the total number of workflows, the number of published workflows, the total number of task instances executed on the day, the task instance success rate, the number of active scheduling rules, and the number of running task instances. The periodic trend visualization unit is used to generate a visualization chart of the execution trend of task instances over the past seven days; The high-frequency workflow ranking unit sorts workflows in descending order of cumulative task instance execution counts, displays the success rate of high-frequency workflows and corresponding task instances, and screens for global abnormal and risky processes; The Recently Executed Instances unit is used to display multiple newly generated task instance records in real time, presenting the execution time, workflow to which the task instance belongs, running status and execution summary, and supporting operation and maintenance personnel to view recent automated execution details and locate sudden abnormal task instances.

9. A low-code software automation task orchestration engine system according to claim 8, characterized in that: When the workflow reaches the manual approval node, the execution link is automatically paused and an approval notification is pushed. After receiving the instruction of manual approval or rejection, the workflow is adaptively resumed or terminated. All approval process data, approval results, and workflow execution data are synchronously stored in the execution monitoring module.

10. A method for implementing a low-code software automation task orchestration engine system as described in any one of claims 1-9, characterized in that, Includes the following steps: S1. The workflow orchestration module responds to the user's visual drag-and-drop operation, calls various atomic components of the atomic component library module to build and generate a DAG structured workflow, completes the parameter configuration and logic verification of each node, publishes and generates the corresponding iteration version of the workflow, and synchronizes the workflow's metadata to the multi-source scheduling trigger module. S2. The multi-source scheduling trigger module receives the metadata of the published workflow and completes the binding association, configures single or multiple trigger conditions for the workflow, turns on the scheduling activation switch, and completes the initial configuration of the trigger method. S3: The multi-source scheduling trigger module polls and monitors various triggering methods in real time. When the triggering conditions are met, it generates the corresponding task instance and execution instruction, and pushes it to the execution monitoring module to wait for execution. S4. The execution monitoring module receives execution instructions and task instance information, loads the structure and parameter configuration of the corresponding workflow, drives each node of the workflow to execute sequentially, records the entire chain operation log and node interaction data, and updates the task instance running status in real time. S5, the execution monitoring module aggregates task instance running data and pushes it to the workbench dashboard module in real time. The workbench dashboard module completes data summarization, indicator statistics, trend visualization and health analysis, realizing closed-loop management of the entire lifecycle of automated task instances from orchestration, scheduling, execution to monitoring iteration.

Images