Distributed industrial cyber-physical system based on shared model tree and working method

Abstract

The application provides a kind of distributed industrial internet of things system based on shared model tree and its working method, it is related to internet of things technical field, including scheduling unit, service cluster unit, service registration center unit, Redis cluster unit, load balancing unit and distributed file service unit, solve the various problems caused by the centralized management of existing industrial internet of things system and the manual centralized backup configuration file, utilize the distributed file service unit of sharing type to combine scheduling unit and each service cluster unit, all services are abstracted into a big tree, and according to the specific load and health status of each service cluster unit monitored by scheduling unit in real time, dynamic advance decision and adjustment are made;Each service cluster unit does not need to interact with scheduling unit directly, only needs to monitor the configuration change in distributed file service unit to separate sub-tree, so as to dynamically adjust the responsibility range of the module responsible for itself.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and in particular to a distributed industrial IoT system and its operating method based on a shared model tree. Background Technology

[0002] With the continuous development of Industry 4.0, data acquisition and information-based control of industrial equipment have become particularly important. In particular, the integration of intelligent mobile robot units, such as AGVs and RGVs, has led to the convergence of logistics, information flow, and process flow. The industrial site is no longer static and transparent, but rather tends towards dynamic multi-flow fusion control.

[0003] As a crucial component supporting upper-level business information systems or scheduling systems, the stability, reliability, and real-time performance of industrial IoT systems often determine the stability of overall production. Current industrial IoT systems are no longer limited to traditional data collection and display; they increasingly need to function as unit systems responsible for local scheduling, analysis, or decision-making.

[0004] Existing industrial IoT system solutions generally adopt centralized management and require manual backup of configuration files. This approach is prone to the following problems:

[0005] 1. Each unit operates independently and is responsible for a specific area. The stability of the overall system depends heavily on the reliability of individual hardware units. This often leads to single points of failure and slow repair response. In particular, when there are AGVs, RGVs, or other units in the system that have high real-time requirements, the entire system may be paralyzed, causing serious losses to production.

[0006] 2. Individual device files are not easy to maintain and are prone to corruption due to hardware or system reasons. A large number of service files often need to be manually backed up, and each change requires a new backup, which is labor-intensive and prone to errors, making the operation and maintenance work extremely complicated.

[0007] 3. Since each unit is manually configured individually, duplicate data collection or omissions between services are likely to occur, making global management inconvenient. In addition, each unit has its own interface, making it inconvenient for upper-layer applications to access it uniformly.

[0008] 4. The overall system cannot monitor the load and health status of each unit in real time, thus making it impossible to dynamically adjust resource allocation in a timely manner to achieve system optimization;

[0009] 5. Each unit operates independently without information sharing. Therefore, decision-making and control between production lines require sharing and transmission through third-party applications, making it difficult to achieve plant-wide decision-making and control. Summary of the Invention

[0010] This invention discloses a distributed industrial IoT system and its working method based on a shared model tree. It solves various problems caused by the centralized management and manual backup of configuration files in existing industrial IoT systems. By using a shared distributed file service unit combined with a scheduling unit and various service cluster units, all services are abstracted into a large tree. Based on the specific load and health status of each service cluster unit monitored in real time by the scheduling unit, dynamic advance decisions and adjustments are made. Each service cluster unit does not need to interact directly with the scheduling unit. It only needs to monitor the configuration changes in the distributed file service unit in real time to extract the tree, thereby dynamically adjusting the scope of responsibility of its own module.

[0011] To achieve the above objectives, the technical solution of the present invention is specifically implemented as follows:

[0012] This invention discloses a distributed industrial IoT system based on a shared model tree, comprising a scheduling unit, a service cluster unit, a service registration center unit, a Redis cluster unit, a load balancing unit, and a distributed file service unit. The scheduling unit monitors the status of the service cluster units in real time and dynamically adjusts the overall device configuration model tree stored in the distributed file service unit based on the status of the service cluster units. Each service cluster unit includes several data acquisition servers, which form a cluster through services. Each data acquisition server connects to a device group. The scheduling unit dynamically allocates a device configuration model subtree to each device group. Each data acquisition server stores the real-time device group data it collects to the Redis cluster unit as a device tag value model subtree. Simultaneously, the system monitors in real time whether the overall device configuration model tree stored in the distributed file service unit has changed. If a change is detected in the distributed file service unit... If the overall device configuration model tree of the meta-storage changes, the corresponding device configuration model subtree is updated; the service registration center unit is used to complete the registration of each data acquisition server in the scheduling unit and service cluster unit; the Redis cluster unit is used to store several device tag value model subtrees in the service cluster unit, and the several device tag value model subtrees form the overall device tag value model tree, and the overall device tag value model tree is updated according to the real-time device group data collected by the several service cluster units; the load balancing unit is used to distribute the load of services in the service cluster unit; the distributed file service unit is used to store the overall device configuration model tree of several service cluster units.

[0013] Furthermore, the same service cluster unit shares a unified device configuration model subtree.

[0014] Furthermore, each data acquisition server in the service cluster unit creates a corresponding device-driven model, edge computing model, and service-driven model based on the device model and according to the configuration attributes.

[0015] Furthermore, the device configuration model tree shared by each of the service cluster units is a subtree of the overall device configuration model tree of the distributed file service unit storage.

[0016] Another aspect of this invention discloses a working method for a distributed industrial IoT system based on a shared model tree, comprising the following steps:

[0017] S1: Real-time device data acquisition and dynamic service adjustment;

[0018] Register all data acquisition servers in the scheduling unit and service cluster unit to the service registration center unit;

[0019] The scheduling unit obtains the service list and assigns responsibility attributes to the overall device configuration model tree of the distributed file service unit storage based on the obtained service list;

[0020] The service cluster unit monitors in real time whether the overall device configuration model tree of the distributed file service unit storage has changed. If a change is detected in the overall device configuration model tree of the distributed file service unit storage, the corresponding regional subtree is updated.

[0021] Each data acquisition server receives its own managed group and establishes corresponding device driver model, edge computing model and service driver model according to the configuration attributes;

[0022] Each data acquisition server connects to its subordinate device groups and stores the real-time collected tag data of each device group in a unified Redis cluster unit;

[0023] If the scheduling unit detects an increase or failure of the data acquisition server in the service cluster unit, it will dynamically adjust the overall device configuration model tree stored in the distributed file service unit, and the corresponding service cluster unit will synchronously update the corresponding device configuration model subtree in real time.

[0024] The scheduling unit monitors the load of each service cluster unit in real time and makes optimization adjustments based on the monitored load status.

[0025] S2: Access by upper-layer applications.

[0026] Furthermore, the access methods for the upper-layer applications include Web API, message middleware, and database.

[0027] Furthermore, the process of upper-layer applications accessing the application via WebAPI includes:

[0028] The scheduling unit is exposed to third-party applications as a gateway;

[0029] When the number of concurrent accesses by uplink applications is low, the scheduling unit responds; when the number of concurrent accesses by uplink applications is high, the scheduling unit responds to a portion of the responses, and the remaining responses are forwarded to the service cluster units for distributed processing.

[0030] Furthermore, the process of upper-layer applications accessing information via message middleware includes:

[0031] Each service cluster unit uses the device ID as a distinguishing identifier to create a Topic, and each publishes it to a unified Broker for external exposure;

[0032] Upper-layer applications subscribe and read data according to the agreed-upon protocol.

[0033] Furthermore, the process of upper-layer applications accessing data via a database includes:

[0034] Each service cluster unit connects to the same database service;

[0035] Establish a unified time-series data service and relational data service for external users.

[0036] Beneficial technical effects:

[0037] 1. This invention discloses a distributed industrial IoT system based on a shared model tree, comprising a scheduling unit, a service cluster unit, a service registration center unit, a Redis cluster unit, a load balancing unit, and a distributed file service unit. It solves various problems arising from the centralized management and manual backup of configuration files required in existing industrial IoT systems. By utilizing a shared distributed file service unit in conjunction with the scheduling unit and each service cluster unit, all services are abstracted into a single tree. Dynamic advance decisions and adjustments are made based on the specific load and health status of each service cluster unit monitored in real time by the scheduling unit. Each service cluster unit does not need to directly interact with the scheduling unit; it only needs to monitor configuration changes in the distributed file service unit in real time to extract the tree, thereby dynamically adjusting the scope of responsibility for its assigned modules.

[0038] 2. In this invention, the distributed file service unit stores the overall device configuration model tree, and the Redis cluster unit stores the overall device tag value model tree. This achieves the separation of values ​​and configurations, minimizes the memory model to the greatest extent, improves the speed of data reading and writing, and forms a unified configuration file, so that the underlying device read and write are discretely distributed while the data values ​​are uniformly summarized.

[0039] 3. In this invention, the forms of access for upper-layer applications include Web API, message middleware, and database. This allows for a flexible response mechanism based on concurrent service requirements, which can be either a direct response at the gateway level or a response distributed downwards.

[0040] 4. In this invention, if the scheduling unit detects an increase or failure of the data acquisition server in the service cluster unit, it dynamically adjusts the overall device configuration model tree stored in the distributed file service unit, and the corresponding service cluster unit updates its corresponding device configuration model subtree. The scheduling unit monitors the load of each service cluster unit in real time and makes adjustments based on the monitored load status, ensuring the dynamic reliability of the overall system, greatly reducing the risk and difficulty of operation and maintenance, and reducing the significant losses caused by local downtime on site. Attached Figure Description

[0041] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0042] Figure 1 This is an overall system block diagram of a distributed industrial IoT system based on a shared model tree, as described in this invention. Detailed Implementation

[0043] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0044] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0045] This invention provides, in one aspect, a distributed industrial IoT system based on a shared model tree, see [link to relevant documentation]. Figure 1 The distributed industrial IoT system based on a shared model tree disclosed in this invention specifically includes a scheduling unit, a service cluster unit, a service registration center unit, a Redis cluster unit, a load balancing unit, and a distributed file service unit. By combining the shared distributed file service unit with the scheduling unit and each service cluster unit, all services are abstracted into a single large tree. Dynamic advance decisions and adjustments are made based on the specific load and health status of each service cluster unit, which is monitored in real time by the scheduling unit. Each service cluster unit does not need to interact directly with the scheduling unit; it only needs to monitor configuration changes in the distributed file service unit in real time to extract the tree, thereby dynamically adjusting the scope of responsibility for its assigned module.

[0046] The following details the various unit modules of the distributed industrial IoT system based on a shared model tree disclosed in this invention:

[0047] Scheduling Unit

[0048] The scheduling unit is used to monitor the status of service cluster units in real time and dynamically adjust the overall device configuration model tree stored in the distributed file service unit according to the status of the service cluster units. That is, the scheduling unit is responsible for monitoring the running status of each service cluster unit and scheduling and controlling the entire system according to the running status of each service cluster unit. For example, when the scheduling unit detects that a service in a service cluster unit has failed, the scheduling unit promptly adjusts the overall device configuration model stored in the distributed file service unit. At the same time, the corresponding service cluster unit also updates the corresponding device configuration model subtree. The scheduling unit also provides a unified interface service call.

[0049] Service Cluster Unit

[0050] Each service cluster unit includes several data acquisition servers. These servers form a cluster through services. For example, a service cluster unit may include Service 1, Service 2, and Service 3. The number of services contained in each service cluster unit is unlimited. (Its main functions include two aspects: 1. Connecting various types of devices, such as various PLCs, sensors, and robots, to the outside world via different physical communication links; 2. Performing edge processing on the collected data according to task instructions or scripts, and then uniformly opening the service to the outside world and providing standard information interfaces to the inside world.) The same service cluster unit shares a unified device configuration model subtree. The device configuration model subtree shared by each service cluster unit is a subtree of the overall device configuration model tree stored by the distributed file service unit. Each data acquisition server connects to a device group. The number of devices in a device group is unlimited. The scheduling unit dynamically allocates a device configuration model subtree to each device group. Each data acquisition server uses the device model as a basis and creates corresponding device driver models, edge computing models, and service driver models based on configuration attributes, and establishes corresponding tag nodes. Each data acquisition server stores the real-time device group data (such as temperature or pressure) in a unified hash format in a Redis cluster unit, which is used as a device tag value model subtree for value sharing. At the same time, the service cluster unit monitors in real time whether the overall device configuration model tree stored in the distributed file service unit has changed. If a change is detected in the overall device configuration model tree stored in the distributed file service unit, the corresponding device configuration model subtree is updated. This ensures that the load of each service is distributed when multiple devices read data at the lower level, while also ensuring unified access for the upper-end application and unified global management of the entire factory's equipment.

[0051] It is important to note that several service clusters are abstracted into a single tree. The addition and modification of devices are managed globally and uniformly. Device IDs are globally unique and cannot be duplicated; otherwise, it may easily lead to signal transmission conflicts in some serial-to-network devices.

[0052] Service Registration Center Unit

[0053] The service registration center unit is used to complete the registration of each data acquisition server in the scheduling unit and service cluster unit.

[0054] Redis cluster unit

[0055] It is used to store the device tag value model subtrees in several service cluster units. The several device tag value model subtrees form the overall device tag value model tree. The overall device tag value model tree is updated according to the real-time device group data collected by several service cluster units, forming an aggregation of device value readings in each device group.

[0056] Load balancing unit

[0057] The load balancing unit is used to distribute the load among the services in the service cluster unit. When the service cluster unit experiences high-frequency and high-concurrency access, the load balancing unit can distribute the pressure among the services in the service cluster unit.

[0058] Distributed file service unit

[0059] The distributed file service unit is used to store the overall device configuration model tree. The device configuration model subtrees of several service cluster units form the overall device configuration model tree, which provides unified monitoring and dynamic extraction for each service and serves as the basis for establishing the driving model of each data acquisition server.

[0060] Another aspect of the present invention discloses a working method for a distributed industrial IoT system based on a shared model tree, comprising the following steps:

[0061] S1: Real-time device data acquisition and dynamic service adjustment, specifically:

[0062] Register all data acquisition servers in the scheduling unit and service cluster unit to the service registration center unit;

[0063] The scheduling unit obtains the service list and assigns responsibility attributes to the overall device configuration model tree of the distributed file service unit storage based on the obtained service list;

[0064] For example, Group1 belongs to Service 1, and Group2 and Group3 belong to Service 2, etc.

[0065] The service cluster unit monitors in real time whether the device configuration model tree stored in the distributed file service unit has changed. If a change is detected in the device configuration model tree stored in the distributed file service unit, the corresponding device configuration model subtree is updated.

[0066] Each data acquisition server receives its own managed group and establishes corresponding device driver model, edge computing model and service driver model according to the configuration attributes;

[0067] Each data acquisition server connects to its subordinate device groups and stores the real-time collected tag data of each device group in a unified Redis cluster unit;

[0068] Specifically, each service connects to the underlying device group, monitors the location of the underlying devices in real time, and when changes occur, they are uniformly summarized into the device tag value model tree in the Redis cluster unit to achieve the aggregation and unification of discrete unit data.

[0069] If the scheduling unit detects an increase or failure of the data acquisition server in the service cluster unit, it will dynamically adjust the overall device configuration model tree stored in the distributed file service unit, and the corresponding service cluster unit will synchronously update the corresponding device configuration model subtree in real time.

[0070] Specifically, if the scheduling unit detects that individual services in the service cluster unit have crashed or increased, it can dynamically adjust the overall device model tree stored in the distributed file service unit according to different strategies such as service distribution and weight allocation, and the corresponding service cluster unit will make rapid adjustments accordingly.

[0071] The scheduling unit monitors the load of each service cluster unit in real time and makes optimization adjustments based on the monitored load status.

[0072] Specifically, the scheduling unit monitors the load and health status of each service in real time, and makes timely dynamic adjustments in case of power outages, memory overruns, or other issues.

[0073] S2: Upper-layer application access

[0074] Specifically, upper-layer applications can access the system via Web API, message middleware, or database.

[0075] When upper-layer applications access via Web API: the scheduling unit acts as a gateway exposed to third-party applications;

[0076] When the number of concurrent accesses by uplink applications is low, the scheduling unit responds, reducing the link path depth. When there are multiple concurrent accesses by uplink applications, intelligent flow selection can be performed based on the application timeliness level and priority. That is, the scheduling unit directly responds to a portion of the requests, and the remaining requests are forwarded to various service cluster units for distributed processing.

[0077] When upper-layer applications access the service using message middleware: each service cluster unit uses the device ID as a distinguishing identifier to create a Topic, and each publishes it to a unified Broker for external exposure; the upper-layer application subscribes and reads the content according to the agreed protocol.

[0078] When upper-layer applications access data via a database: each service cluster unit connects to the same database service; and a unified time-series data service and relational data service are established for external use.

[0079] This invention discloses a distributed industrial IoT system and its working method based on a shared model tree. It utilizes a shared distributed file service unit and a scheduling unit to abstract all services into a unified tree. The scheduling unit monitors the specific load and health status of each service in real time, making dynamic advance decisions and adjustments. Each service cluster unit does not need to directly interact with the scheduling unit; it only needs to monitor the configuration changes of the overall device configuration model tree stored in the distributed file service unit to extract the tree, thereby dynamically adjusting the scope of responsibility for its assigned module. This ensures unified management of each service cluster unit and effectively solves the problems of previous independent system deployments, inability to share resources in a timely manner, inconvenient configuration and maintenance, and susceptibility to single points of failure. Simultaneously, to improve the collaboration between unit modules, the system extracts the overall device configuration model tree stored in the distributed file service unit... The device, node, and related value attributes are stored as a device tag value model tree in the Redis cluster unit. This achieves separation of values ​​and configurations, minimizes the memory model, improves data read and write speed, and forms a highly efficient, loosely coupled architecture where configuration files are unified, underlying device read and write operations are discretely distributed, and data values ​​are uniformly aggregated. The overall system, with the help of the scheduling unit, can adopt a flexible response mechanism based on concurrent service requirements, either by responding directly at the gateway level or by distributing the pressure downwards. At the device layer, there is a highly reliable system to ensure the stability of the system configuration files. At the same time, proactive intervention is performed in advance to prevent single-machine failures or suboptimal health to maintain dynamic balance. This ensures the dynamic reliability of the overall system, greatly reduces the risk and difficulty of operation and maintenance, and minimizes significant losses caused by partial downtime.

[0080] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0081] The above embodiments are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A distributed industrial IoT system based on a shared model tree, characterized in that, include: The scheduling unit is used to monitor the status of the service cluster unit in real time and dynamically adjust the overall device configuration model tree stored in the distributed file service unit according to the status of the service cluster unit. The service cluster unit includes several data acquisition servers. The several data acquisition servers form a cluster through services. Each data acquisition server is connected to a device group. The scheduling unit dynamically allocates a device configuration model subtree to each device group. Each data acquisition server stores the collected real-time device group data to the Redis cluster unit as a device tag value model subtree. It also monitors in real time whether the overall device configuration model tree of the distributed file service unit storage has changed. If a change is detected in the overall device configuration model tree of the distributed file service unit storage, the corresponding device configuration model subtree is updated. The service registration center unit is used to complete the registration of each data acquisition server in the scheduling unit and service cluster unit; A Redis cluster unit is used to store several device tag value model subtrees in the service cluster unit. The several device tag value model subtrees form an overall device tag value model tree, and the overall device tag value model tree is updated according to the real-time device group data collected by the several service cluster units. A load balancing unit is used to distribute the load among the services in the service cluster unit. A distributed file service unit is used to store the overall device configuration model tree of several of the service cluster units; The same service cluster unit shares a unified device configuration model subtree; Each data acquisition server in the service cluster unit creates a corresponding device-driven model, edge computing model, and service-driven model based on the device model and configuration attributes. The device configuration model tree shared by each of the service cluster units is a subtree of the overall device configuration model tree of the distributed file service unit storage.

2. The working method of a distributed industrial IoT system based on a shared model tree according to claim 1, characterized in that, Includes the following steps: S1: Real-time device data acquisition and dynamic service adjustment; Register all data acquisition servers in the scheduling unit and service cluster unit to the service registration center unit; The scheduling unit obtains the service list and assigns responsibility attributes to the overall device configuration model tree of the distributed file service unit storage based on the obtained service list; The service cluster unit monitors in real time whether the overall device configuration model tree of the distributed file service unit storage has changed. If a change is detected in the overall device configuration model tree of the distributed file service unit storage, the corresponding device configuration model subtree is updated. Each data acquisition server receives its own managed group and establishes corresponding device driver model, edge computing model and service driver model according to the configuration attributes; Each data acquisition server connects to its subordinate device groups and stores the real-time collected tag data of each device group in a unified Redis cluster unit; If the scheduling unit detects an increase or failure of the data acquisition server in the service cluster unit, it will dynamically adjust the overall device configuration model tree stored in the distributed file service unit, and the corresponding service cluster unit will synchronously update the corresponding device configuration model subtree in real time. The scheduling unit monitors the load of each service cluster unit in real time and makes optimization adjustments based on the monitored load status; S2: Access by upper-layer applications.

3. The working method of a distributed industrial IoT system based on a shared model tree according to claim 2, characterized in that, The upper-layer application access methods include Web API, message middleware, and database.

4. The working method of a distributed industrial IoT system based on a shared model tree according to claim 3, characterized in that, The process of upper-layer applications accessing the web via API includes: The scheduling unit is exposed to third-party applications as a gateway; When the number of concurrent accesses by uplink applications is low, the scheduling unit responds directly; when there are multiple concurrent accesses by uplink applications, the scheduling unit responds to a portion of the responses, and the remaining responses are forwarded to the service cluster units for distributed processing.

5. The working method of a distributed industrial IoT system based on a shared model tree according to claim 3, characterized in that, The process of upper-layer applications accessing information via message middleware includes: Each service cluster unit uses the device ID as a distinguishing identifier to create a Topic, and each publishes it to a unified Broker for external exposure; Upper-layer applications subscribe and read data according to the agreed-upon protocol.

6. The working method of a distributed industrial IoT system based on a shared model tree according to claim 3, characterized in that, The process of upper-layer applications accessing databases includes: Each service cluster unit connects to the same database service; Establish a unified time-series data service and relational data service for external users.

Images