A high-precision distributed intelligent control platform

Abstract

The present application provides a high-precision distributed intelligent control platform, which relates to the field of intelligent control technology. The distributed intelligent control platform includes a cloud layer, an edge layer and an end layer; the end layer is used to access various devices of the system to be controlled; the edge layer is used to provide multiple functional modules for modular control of various devices of the system to be controlled, and to generate real-time control instructions based on multiple functional modules and data of various devices of the system to be controlled, and then to perform real-time control of various devices of the system to be controlled according to the real-time control instructions; the cloud layer is used to receive the processing request of the first type of task sent by the edge layer, process the first type of task, and return the processing result of the first type of task to the edge layer; the cloud layer is also used to send the second type of task to the edge layer, and receive the processing result of the second type of task reported by the edge layer. This embodiment can meet the requirements of wide-area spatial distribution, real-time, low latency and data privacy.

Description

Technical Field

[0001] The present application relates to the field of intelligent control technology, and in particular to a high-precision distributed intelligent control platform. Background Art

[0002] With the rapid development of science and technology, intelligent control systems play an increasingly important role in industrial automation, smart city construction, environmental monitoring and other fields. However, existing technologies still face many challenges in achieving high-precision control and wide-area spatial distribution. For example, traditional control systems are often limited to local areas and it is difficult to achieve large-scale, cross-regional real-time and precise control. Therefore, this technical problem needs to be solved urgently. Summary of the invention

[0003] In view of the above problems, this application is proposed to provide a high-precision distributed intelligent control platform that overcomes the above problems or at least partially solves the above problems, and has the characteristics of time-series wide-area spatial distribution. The technical solution is as follows:

[0004] The embodiment of the present application provides a high-precision distributed intelligent control platform, the distributed intelligent control platform comprising: a cloud layer, an edge layer and a terminal layer;

[0005] The end layer is used to access each device of the system to be controlled, collect data of each device of the system to be controlled, and send the data of each device of the system to be controlled to the edge layer; wherein each device of the system to be controlled is distributed in a wide area;

[0006] The edge layer is used to provide multiple functional modules for modular control of each device of the controlled system, generate real-time control instructions based on the multiple functional modules and the data of each device of the controlled system, and then perform real-time control of each device of the controlled system according to the real-time control instructions; and assume the pre-processing function of the data uploaded by the end layer to reduce the data transmission volume directly between the cloud layer and the edge layer;

[0007] The edge layer is further used to send a processing request for a first type of task to the cloud layer, and receive a processing result returned by the cloud layer according to the processing request for the first type of task; wherein the first type of task is characterized by a large amount of processing data, a long processing time, and a processing result covering the global control planning of multiple control modules of the edge layer;

[0008] The edge layer is further used to receive and execute the second type of tasks issued by the cloud layer, and report the processing results of the second type of tasks to the cloud layer; wherein the second type of tasks are tasks characterized by centralized planning by the cloud brain and distributed implementation by the edge;

[0009] The edge layer is also used to provide lightweight computing, analysis, processing and decision-making;

[0010] The cloud layer is used to receive the processing request of the first type of task sent by the edge layer, process the first type of task, obtain the processing result of the first type of task, and return the processing result of the first type of task to the edge layer;

[0011] The cloud layer is further used to send the second type of tasks to the edge layer and receive the processing results of the second type of tasks reported by the edge layer.

[0012] In a possible implementation, taking an accelerator as an example, the system to be controlled includes an accelerator, and the various devices of the accelerator connected to the end layer include one or more of a charged particle generating device, a vacuum acceleration chamber, a beam transport system, a beam monitoring and diagnostic device, an electromagnetic field stabilization control device, and a power supply and operating device.

[0013] In a possible implementation, the cloud layer includes cloud brain, cloud storage, cloud computing and cloud security;

[0014] The cloud brain is used to manage and coordinate computing, storage and network resources within the cloud layer;

[0015] The cloud storage is used to store data and provide data support for cloud computing;

[0016] The cloud computing is used to process and analyze data;

[0017] The cloud security is used to provide security capabilities for computing, storage and network resources within the cloud layer.

[0018] In a possible implementation, the cloud layer includes, but is not limited to, a mode management unit, a resource scheduling unit, an alarm unit, a mirror unit, various types of databases, an operation status analysis unit, a simulation and emulation unit, an artificial intelligence unit, a visualization unit, an intelligent operation and maintenance unit, and a security management unit;

[0019] The mode management unit is used to configure and manage the mode according to the plan and objectives of the operation of the system to be controlled, and to issue and allocate all tasks under the configured mode;

[0020] The resource scheduling unit is used to dynamically and flexibly schedule and use computing, storage and network resources according to various task requirements;

[0021] The alarm unit is used to push sound, light or SMS messages according to the set strategy and priority of the device status of the abnormal device according to the analysis results of the operating status of each device in the system to be controlled;

[0022] The mirror unit is used to package various standard or customized systems, services, and software resources into file resources through containerization and virtualization, so that the intelligent operation and maintenance unit can retrieve, deploy, and perform intelligent installation and configuration;

[0023] The various types of databases include one or more of an artificial intelligence model library, a visualization component library, a function library, a document library, and a configuration parameter library;

[0024] The operation status analysis unit is used to monitor and analyze the operation parameters of each device, software service and device of the controlled system;

[0025] The simulation and emulation unit is used to simulate and emulate the execution process of the cloud layer;

[0026] The artificial intelligence unit is used for one or more of fault prediction, fault analysis, health diagnosis, and automatic beam adjustment;

[0027] The visualization unit is used to display the filtered and analyzed data based on the visualization component library;

[0028] The intelligent operation and maintenance unit is used to perform intelligent operation and maintenance management on the hardware devices and virtual machines involved in the cloud-edge-device architecture;

[0029] The security management unit is used to provide security capabilities for computing, storage and network resources within the cloud layer.

[0030] In a possible implementation, the cloud layer further includes a user interaction unit for receiving a plan and a goal of the operation of the system to be controlled submitted by a user.

[0031] In a possible implementation, the edge layer includes a number of edge nodes, a shadow implementation unit, a message management unit, a real-time calculation unit, a snapshot storage unit, and a regional decision unit;

[0032] The shadow implementation unit is used to copy the service mappings of the cloud layer to the edge node;

[0033] The message management unit is used to provide message communication and collaboration mechanisms for different modules and services;

[0034] The real-time calculation unit is used to generate real-time control instructions based on the data of multiple functional modules and various devices of the system to be controlled;

[0035] The snapshot storage unit is used to temporarily store the snapshot generated by the edge layer;

[0036] The regional decision unit is used to perform regional decision analysis and processing on the second type of tasks sent by the cloud layer.

[0037] In a possible implementation, the edge layer is used to provide the underlying hardware of multiple functional modules for modularly controlling each device of the system to be controlled, and is a board card for implementing a configurable control function.

[0038] In a possible implementation, the board includes a baseboard, a core board, and a functional sub-board;

[0039] The core board is pluggably connected to the base board;

[0040] The functional sub-board is pluggably connected to the base board.

[0041] In a possible implementation, the core board is implemented based on a field programmable gate array (FPGA).

[0042] In a possible implementation, the control platform includes a data network and a real-time network;

[0043] The data network is used for transmitting various types of data of the cloud layer, the edge layer and the terminal layer;

[0044] The real-time network is used for timing triggering, time synchronization and transmission of real-time data;

[0045] The data network and the real-time network are constructed by wired or wireless means, with different levels of time accuracy, providing time references from ns, us, ms to s, and completing intelligent control according to strict logical timing.

[0046] By means of the above technical scheme, the high-precision distributed intelligent control platform provided in the embodiment of the present application includes a cloud layer, an edge layer and an end layer; the end layer is used to access each device of the system to be controlled, collect data of each device of the system to be controlled, and send the data of each device of the system to be controlled to the edge layer; the edge layer is used to provide multiple functional modules for modular control of each device of the system to be controlled, generate real-time control instructions based on the multiple functional modules and the data of each device of the system to be controlled, and then perform real-time control on each device of the system to be controlled according to the real-time control instructions; the edge layer is also used to send data to the cloud The edge layer sends a processing request for a first type of task, and receives the processing result returned by the cloud layer according to the processing request for the first type of task; the edge layer is also used to receive and execute the second type of task sent by the cloud layer, and report the processing result of the second type of task to the cloud layer; the cloud layer is used to receive the processing request for the first type of task sent by the edge layer, process the first type of task, obtain the processing result of the first type of task, and return the processing result of the first type of task to the edge layer; the cloud layer is also used to send the second type of task to the edge layer, and receive the processing result of the second type of task reported by the edge layer. It can be seen that this embodiment adopts the cloud-edge-end architecture as a centralized and distributed computing model, combines cloud computing, edge computing and local computing, and allocates computing resources from the traditional centralized cloud to the network edge and various devices of the system to be controlled, which can better meet the requirements of wide-area spatial distribution, real-time, low latency and data privacy. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in describing the embodiments of the present application are briefly introduced below.

[0048] Figure 1 The structure diagram of the high-precision distributed intelligent control platform provided by the embodiment of the present application is shown;

[0049] Figure 2 A structural diagram of a high-precision distributed intelligent control platform taking an accelerator as an example provided by another embodiment of the present application is shown. DETAILED DESCRIPTION

[0050] The exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0051] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such use is interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the term "including" and its variants are to be interpreted as open-ended terms meaning "including but not limited to".

[0052] The embodiment of the present application provides a high-precision distributed intelligent control platform, which has the characteristics of time-series wide-area spatial distribution and can control the controlled system in real time, efficiently and safely. The distributed intelligent control platform has the following features:

[0053] 1) The equipment of the system to be controlled can be widely distributed and can be distributed in different geographical locations;

[0054] 2) If high-precision logical synchronization timing is required, time bases of ns (nanoseconds), us (microseconds), ms (milliseconds) to s (seconds) can be provided;

[0055] 3) Strict logical timing can be required;

[0056] 4) Prediction and real-time feedback work together;

[0057] 5) In terms of artificial intelligence, it has intelligent processing of perception, decision-making and execution.

[0058] The distributed intelligent control platform is applicable to both high logic and time requirements. The system to be controlled can be a large-scale distributed synchronous real-time control system, a wide-area distributed high-synchronization real-time control system, such as an accelerator, a reactor, a fission device, a fusion device, chemical industry, electric power, a large accelerator, etc., which is not limited in this embodiment.

[0059] The system to be controlled may also be a system that has relatively high time accuracy requirements for the logical sequence of operations, such as a closely related production line.

[0060] The distributed intelligent control platform executes the model, collects data from the terminal layer for training on the cloud, and then sends it down for execution, which is an artificial intelligence entity.

[0061] like Figure 1 As shown, the high-precision distributed intelligent control platform may include: a cloud layer, an edge layer, and a terminal layer;

[0062] The end layer is used to access each device of the system to be controlled, collect data of each device of the system to be controlled, and send the data of each device of the system to be controlled to the edge layer; wherein each device of the system to be controlled is distributed in a wide area;

[0063] The edge layer is used to provide multiple functional modules for modular control of each device of the controlled system, generate real-time control instructions based on the multiple functional modules and the data of each device of the controlled system, and then perform real-time control of each device of the controlled system according to the real-time control instructions; and assume the pre-processing function of the data uploaded by the end layer to reduce the data transmission volume directly between the cloud layer and the edge layer;

[0064] The edge layer is further used to send a processing request for a first type of task to the cloud layer, and receive a processing result returned by the cloud layer according to the processing request for the first type of task; wherein the first type of task is characterized by a large amount of processing data, a long processing time, and a processing result covering the global control planning of multiple control modules of the edge layer;

[0065] The edge layer is further used to receive and execute the second type of tasks issued by the cloud layer, and report the processing results of the second type of tasks to the cloud layer; wherein the second type of tasks are tasks characterized by centralized planning by the cloud brain and distributed implementation by the edge;

[0066] The edge layer is also used to provide lightweight computing, analysis, processing and decision-making;

[0067] The cloud layer is used to receive the processing request of the first type of task sent by the edge layer, process the first type of task, obtain the processing result of the first type of task, and return the processing result of the first type of task to the edge layer;

[0068] The cloud layer is further used to send the second type of tasks to the edge layer and receive the processing results of the second type of tasks reported by the edge layer.

[0069] This embodiment adopts a cloud-edge-end architecture as a centralized and distributed computing model, combining cloud computing, edge computing, and local computing, and allocating computing resources from the traditional centralized cloud to the network edge and various devices in the controlled system, which can better meet the needs of wide-area spatial distribution, real-time, low latency, and data privacy.

[0070] A possible implementation method is provided in an embodiment of the present application. Taking an accelerator as an example, the system to be controlled may include an accelerator, and the various devices of the accelerator connected to the end layer may include a charged particle generating device, a vacuum acceleration chamber, a beam transport system, a beam monitoring and diagnostic device, an electromagnetic field stability control device, and one or more of a power supply and operating device. This embodiment does not limit this.

[0071] Taking a superconducting proton linear accelerator as an example, the various devices of the superconducting proton linear accelerator may include a room temperature front end accelerator, a superconducting linear accelerator, a high energy transmission line and a beam collection terminal, wherein the room temperature front end accelerator may include an ion source and a low energy transmission section, a radio frequency quadrupole accelerator, and a medium energy transmission section; the superconducting linear accelerator may include a superconducting half-wavelength cavity acceleration section, a superconducting spoke cavity acceleration section, and a superconducting ellipsoid cavity acceleration section; the high energy transmission line and the beam collection terminal may include a high energy transmission section and a beam collection terminal, etc. The key hardware systems involved here may include a magnet system, a power supply system, a vacuum system, a beam diagnosis system, a high frequency power, a superconducting radio frequency, a room temperature radio frequency, a cryostat, an ion source, etc., and this embodiment does not limit this.

[0072] A possible implementation method is provided in an embodiment of the present application, wherein the cloud layer may include cloud brain, cloud storage, cloud computing and cloud security;

[0073] The cloud brain is used to manage and coordinate computing, storage and network resources within the cloud layer;

[0074] The cloud storage is used to store data and provide data support for cloud computing;

[0075] The cloud computing is used to process and analyze data;

[0076] The cloud security is used to provide security capabilities for computing, storage and network resources within the cloud layer.

[0077] A possible implementation method is provided in an embodiment of the present application, wherein the cloud layer may include but is not limited to a mode management unit, a resource scheduling unit, an alarm unit, a mirror unit, various types of databases, an operation status analysis unit, a simulation and emulation unit, an artificial intelligence unit, a visualization unit, an intelligent operation and maintenance unit, and a security management unit;

[0078] The mode management unit is used to configure and manage the mode according to the plan and objectives of the operation of the system to be controlled, and to issue and allocate all tasks under the configured mode;

[0079] The resource scheduling unit is used to dynamically and flexibly schedule and use computing, storage and network resources according to various task requirements;

[0080] The alarm unit is used to push sound, light or SMS messages according to the set strategy and priority of the device status of the abnormal device according to the analysis results of the operating status of each device in the system to be controlled;

[0081] The mirror unit is used to package various standard or customized systems, services, and software resources into file resources through containerization and virtualization, so that the intelligent operation and maintenance unit can retrieve, deploy, and perform intelligent installation and configuration;

[0082] The various types of databases include but are not limited to one or more of an artificial intelligence model library, a visualization component library, a function library, a document library, and a configuration parameter library;

[0083] The operation status analysis unit is used to monitor and analyze the operation parameters of each device, software service and device of the controlled system;

[0084] The simulation and emulation unit is used to simulate and emulate the execution process of the cloud layer;

[0085] The artificial intelligence unit is used for one or more of fault prediction, fault analysis, health diagnosis, and automatic beam adjustment;

[0086] The visualization unit is used to display the filtered and analyzed data based on the visualization component library;

[0087] The intelligent operation and maintenance unit is used to perform intelligent operation and maintenance management on the hardware devices and virtual machines involved in the cloud-edge-device architecture;

[0088] The security management unit is used to provide security capabilities for computing, storage and network resources within the cloud layer.

[0089] A possible implementation method is provided in an embodiment of the present application, and the cloud layer may further include a user interaction unit for receiving plans and goals for operation of the system to be controlled submitted by a user.

[0090] A possible implementation method is provided in an embodiment of the present application, wherein the edge layer may include but is not limited to a number of edge nodes, a shadow implementation unit, a message management unit, a real-time calculation unit, a snapshot storage unit, and a regional decision unit;

[0091] The shadow implementation unit is used to copy the service mappings of the cloud layer to the edge node;

[0092] The message management unit is used to provide message communication and collaboration mechanisms for different modules and services;

[0093] The real-time calculation unit is used to generate real-time control instructions based on the data of multiple functional modules and various devices of the system to be controlled;

[0094] The snapshot storage unit is used to temporarily store the snapshot generated by the edge layer;

[0095] The regional decision unit is used to perform regional decision analysis and processing on the second type of tasks sent by the cloud layer.

[0096] A possible implementation method is provided in an embodiment of the present application, wherein the edge layer is used to provide the underlying hardware of multiple functional modules for modularly controlling each device of the controlled system, and is a board for implementing a configurable control function.

[0097] A possible implementation method is provided in an embodiment of the present application, wherein the board may include a baseboard, a core board, and a functional sub-board;

[0098] The core board is pluggably connected to the base board;

[0099] The functional sub-board is pluggably connected to the base board.

[0100] In this embodiment, the core board can run the operating system, process and store data, build software algorithms, etc., and the software algorithms can realize modular configurable functions. The functional sub-boards are mainly designed independently to meet different industrial control scenarios, and at the same time meet the EMC (Electro Magnetic Compatibility) standards in industrial control systems.

[0101] A possible implementation method is provided in the embodiment of the present application, and the core board is implemented based on FPGA (Field-Programmable Gate Array). Not only can FPGA digital hardware logic be implemented, but also an onboard system can be carried, generally using a Linux system, which builds related IOC (Inversion of Control) and can perform corresponding software algorithms, and the software algorithm has also realized modular configurable functions.

[0102] In the above embodiments, the board can not only be installed in an independent chassis, but also can be installed in multiple chassis. It is suitable for industrial control chassis such as CPCI (CompactPCI, a computer bus interface standard) and VXI (VMEbus eXtension, an expansion bus standard based on VME bus technology), and realizes different structures of assembly and single installation.

[0103] A possible implementation method is provided in an embodiment of the present application, wherein the control platform may include a data network and a real-time network;

[0104] The data network is used for transmitting various types of data of the cloud layer, the edge layer and the terminal layer;

[0105] The real-time network is used for timing triggering, time synchronization and transmission of real-time data.

[0106] The above introduces Figure 1There are various implementation methods for each link of the illustrated embodiment. The high-precision distributed intelligent control platform with time-series wide-area spatial distribution of the embodiment of the present application will be further explained below through specific embodiments.

[0107] Take the accelerator as an example. The accelerator, also called particle accelerator or charged particle accelerator, is a device that uses electromagnetic fields to confine microscopic charged particles in a specific area (mostly a vacuum area) and accelerate them to reach higher energy. Accelerators can accelerate charged particles such as electrons, protons, and ions, so that the speed of particles reaches several thousand kilometers per second, tens of thousands of kilometers per second, or even close to the speed of light, and make particles reach higher energy. For example, an electron accelerator injects electrons with an energy of 20 GeV (gigaelectron volts), and the corresponding electron speed is 0.99999999979 times the speed of light; after the electrons are accelerated by the accelerator, the energy can reach 100 GeV, and the electron speed reaches 0.9999999999987 times the speed of light.

[0108] The control system is the brain and nerves of the accelerator operation. The main task of the control system is to automatically monitor and control the working status of each device of the accelerator operating in high-power, high-current, and multi-beam modes, and provide operators with equipment operating parameters and beam adjustment methods to coordinate the operation of various devices to obtain the ideal beam current. The operating status of each device of the accelerator is closely related to the real-time, stability, and reliability of the control system. Therefore, how to provide a real-time, stable, and reliable control system has become a technical problem that needs to be solved urgently.

[0109] Data processing and data-based intelligent services are becoming increasingly important in the IoT architecture for the next-generation information infrastructure. With the rapid development of cloud computing and edge computing, the cloud-edge-end architecture, as a centralized and distributed computing model, can meet the needs of CiADS (China initiative Accelerator Driven System) to build a control system. This embodiment proposes a design scheme based on the cloud-edge-end architecture to achieve reliable, efficient, and secure computing and data processing services.

[0110] like Figure 2 As shown in the figure, the Cloud layer is responsible for managing and allocating computing resources and taking on large-scale data processing and storage tasks; the Edge layer provides closer computing and data processing capabilities; and the End layer accesses various devices of the accelerator.

[0111] Cloud, Edge and End are the contents of the "cloud, edge and end" structure. For entities, Cloud, as the central cloud, is responsible for overall decision-making, computing and storage tasks, and sends instructions to Edge through the cloud brain; Edge can be a powerful GIANT (general intelligent control front end) device on site, or an edge server with relatively stronger computing or storage capabilities connected to all front-end control devices in a certain system of the accelerator. GIANT devices all have powerful and intelligent general core boards, and their service objects can be selected according to system requirements, that is, End in the figure. End can be a magnet power supply, vacuum instrument, beam diagnostic element, low-level node or various front-end sensors and other devices.

[0112] exist Figure 2 In the present invention, Cloud may include decision-making, storage, analysis and cloud management, etc. Decision-making may include Beam Operating System (BOS), mode management, resource scheduling, alarm, etc.; storage may include archiving, mirroring, logs and libraries; analysis may include visualization, simulation and emulation, operation status analysis and AI (Artificial Intelligence); cloud management may include security management, intelligent operation and maintenance, etc., which are not limited in this embodiment.

[0113] Edge may include real-time computing, regional decision-making, shadow implementation, data preprocessing, functional safety, snapshot storage, message management, self-running, analytical execution, embedded AI, data collection, time synchronization, equipment control, and interlocking / protection, etc., which are not limited in this embodiment.

[0114] End can be a sensor, flow meter, camera, probe, magnet power supply, vacuum instrument, beam diagnostic element, low-level node or various front-end sensors and other devices.

[0115] exist Figure 2 In the embodiment, the control platform may include a data network and a real-time network; the data network is used for the transmission of various types of data of Cloud, Edge and End; the real-time network is used for timing triggering, time synchronization and transmission of real-time data; wherein the data network and the real-time network are established in a wired or wireless manner, with different levels of time accuracy, providing time references from ns, us, ms to s, and completing intelligent control according to a strict logical sequence.

[0116] Furthermore, Cloud may include a user interaction unit, a mode management unit, a resource scheduling unit, an alarm unit, a mirroring unit, various types of databases, an operation status analysis unit, a simulation and emulation unit, an artificial intelligence unit, a visualization unit, an intelligent operation and maintenance unit, and a security management unit;

[0117] The user interaction unit is used to receive the accelerator operation plan and goal submitted by the user;

[0118] The mode management unit is used to configure and manage the mode according to the plan and goal of the accelerator operation, and to issue and allocate all tasks in the configured mode;

[0119] The resource scheduling unit is used to dynamically and flexibly schedule and use computing, storage and network resources according to various task requirements;

[0120] The alarm unit is used to push sound, light or SMS messages according to the set strategy and priority of the device status of the abnormal operation according to the analysis results of the operation status of each device of the accelerator;

[0121] The mirror unit is used to package various standard or customized systems, services, and software resources into file resources through containerization and virtualization, so that the intelligent operation and maintenance unit can retrieve, deploy, and perform intelligent installation and configuration;

[0122] The various types of databases include one or more of an artificial intelligence model library, a visualization component library, a function library, a document library, and a configuration parameter library;

[0123] The operation status analysis unit is used to monitor and analyze the operation parameters of various devices, software services and devices of the accelerator;

[0124] The simulation and emulation unit is used to simulate and emulate the execution process of the cloud layer;

[0125] The artificial intelligence unit is used for one or more of fault prediction, fault analysis, health diagnosis, and automatic beam adjustment;

[0126] The visualization unit is used to display the filtered and analyzed data based on the visualization component library;

[0127] The intelligent operation and maintenance unit is used to perform intelligent operation and maintenance management on the hardware devices and virtual machines involved in the cloud-edge-device architecture;

[0128] The security management unit is used to provide security capabilities for computing, storage and network resources within the cloud layer.

[0129] Edge can include several edge nodes, shadow implementation units, message management units, real-time computing units, snapshot storage units, and regional decision units;

[0130] The shadow implementation unit is used to copy the service mappings of the cloud layer to the edge node;

[0131] The message management unit is used to provide message communication and collaboration mechanisms for different modules and services;

[0132] The real-time calculation unit is used to generate real-time control instructions based on the data of the multiple functional modules and each device of the accelerator;

[0133] The snapshot storage unit is used to temporarily store the snapshot generated by the edge layer;

[0134] The regional decision unit is used to perform regional decision analysis and processing on the second type of tasks sent by the cloud layer.

[0135] The underlying hardware of the Edge for providing multiple functional modules for modular control of various devices of the accelerator is a board for implementing configurable control functions. Furthermore, the board may include a baseboard, a core board, and a functional sub-board; the core board is pluggable and connected to the baseboard; the functional sub-board is pluggable and connected to the baseboard. The core board is implemented based on FPGA.

[0136] The cloud-edge architecture provided in this embodiment is flexible and can automatically adjust computing resource allocation and data processing capabilities according to demand.

[0137] Elastic computing resource allocation: The cloud-edge architecture should be elastic and be able to automatically adjust the allocation of computing resources according to demand. This means that during peak load periods, the architecture can dynamically allocate more computing resources to meet the needs of high concurrency and large-scale data processing; and during low load periods, it can automatically release excess computing resources to save costs.

[0138] Automatic expansion and contraction: The cloud-edge architecture supports automatic expansion and contraction. When more computing resources are needed, the architecture can automatically expand computing nodes or increase server instances to improve computing power. When the load decreases, the architecture can also automatically shrink computing nodes or stop server instances that are no longer needed to save resources and costs.

[0139] Horizontal expansion and vertical expansion: The cloud-edge architecture can improve flexibility and scalability through horizontal expansion and vertical expansion. Horizontal expansion refers to adding more computing nodes or server instances to increase parallel processing capabilities. Vertical expansion increases computing power by improving the performance of a single computing node or server instance.

[0140] Distributed data processing: The cloud-edge architecture can distribute data processing tasks to the cloud, edge nodes, and accelerator devices to achieve distributed data processing capabilities. This can fully utilize the computing resources of each node and improve the processing power and efficiency of the entire system.

[0141] Service containerization: Containerizing services in the cloud-edge architecture can improve flexibility and scalability. By packaging different services into independent containers, rapid deployment, expansion, and migration can be achieved. When there is a new service demand, you only need to start the corresponding container without reconfiguring the entire system.

[0142] Through the above design and implementation methods, the cloud-edge architecture can better meet the needs of different application scenarios and provide flexible and scalable computing and data processing services.

[0143] In addition, security is an important consideration in the design of cloud-edge architecture, including data encryption, access control, and protection mechanisms.

[0144] Data isolation and encryption: The cloud-edge architecture uses data isolation and encryption technology to ensure that data is not accessed or stolen by unauthorized persons during transmission and storage. This includes encrypting data and taking strict access control and permission management measures during data transmission and storage.

[0145] Device security: The cloud-edge architecture focuses on device security, and ensures that the device is not subject to malicious attacks and tampering during startup and operation through technical means such as secure hardware modules, secure boot and firmware verification.

[0146] Identity authentication and access control: The cloud-edge architecture verifies the identities of users and devices through identity authentication and access control mechanisms, and limits their access rights to prevent unauthorized access and operations.

[0147] Security monitoring and response: The cloud-edge architecture uses a security monitoring and response system to monitor the security status of devices and networks in real time, promptly detect and respond to security threats and attacks, and ensure the safe and stable operation of the system.

[0148] In general, the cloud-edge architecture ensures the security of data and devices through multi-layer security protection and technical means, ensuring that the system can effectively respond to various security threats and risks, thereby improving overall security and stability.

[0149] Furthermore, considering the diversity of different devices and network environments of accelerators, the architecture supports interoperability with various hardware and software platforms.

[0150] Furthermore, since the characteristic of the cloud-edge architecture is to place computing and data processing close to the accelerator devices, it is necessary to ensure low-latency real-time performance.

[0151] Edge computing: The cloud-edge architecture deploys computing resources and data storage at edge nodes that are closer to users or devices, allowing data processing and computing to be completed at closer distances, thereby reducing the time and delay of data transmission.

[0152] Fast data processing: The cloud-edge architecture uses distributed computing and fast data processing technologies to achieve rapid processing and analysis of large-scale data, thereby reducing data processing delays.

[0153] Efficient communication protocol: The cloud-edge architecture adopts efficient communication protocols and data transmission technologies, which can achieve fast data transmission and communication, thereby reducing data transmission delays.

[0154] Optimized algorithms and models: The cloud-edge architecture uses optimized algorithms and models to enable rapid processing and analysis of data, thereby reducing data processing latency.

[0155] To sum up, the low latency of the cloud-edge architecture is reflected in the speed and efficiency of data processing and communication, which is achieved through technical means such as edge computing, fast data processing, efficient communication protocols, and optimization algorithms and models.

[0156] Those skilled in the art can clearly understand that the specific working processes of the systems, devices, and modules described above can refer to the corresponding processes in the aforementioned method embodiments, and for the sake of brevity, they are not further described here.

[0157] Those skilled in the art can understand that the technical solution of the present application can be essentially or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, which includes a number of program instructions, so that an electronic device (such as a personal computer, a server, or a network device, etc.) executes all or part of the steps of the method described in each embodiment of the present application when running the program instructions. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0158] Alternatively, all or part of the steps of implementing the aforementioned method embodiments may be accomplished by hardware associated with program instructions (such as electronic devices such as personal computers, servers, or network devices), and the program instructions may be stored in a computer-readable storage medium. When the program instructions are executed by a processor of an electronic device, the electronic device executes all or part of the steps of the methods described in the embodiments of the present application.

[0159] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that within the spirit and principles of the present application, they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate from the protection scope of the present application.

Claims

1. A high-precision distributed intelligent control platform, characterized in that: The distributed intelligent control platform includes: a cloud layer, an edge layer and a terminal layer; The end layer is used to access each device of the system to be controlled, collect data of each device of the system to be controlled, and send the data of each device of the system to be controlled to the edge layer; wherein each device of the system to be controlled is distributed in a wide area; The edge layer is used to provide multiple functional modules for modular control of each device of the controlled system, generate real-time control instructions based on the multiple functional modules and the data of each device of the controlled system, and then perform real-time control of each device of the controlled system according to the real-time control instructions; and assume the pre-processing function of the data uploaded by the end layer to reduce the data transmission volume directly between the cloud layer and the edge layer; The edge layer is further used to send a processing request for a first type of task to the cloud layer, and receive a processing result returned by the cloud layer according to the processing request for the first type of task; wherein the first type of task is characterized by a large amount of processing data, a long processing time, and a processing result covering the global control planning of multiple control modules of the edge layer; The edge layer is further used to receive and execute the second type of tasks issued by the cloud layer, and report the processing results of the second type of tasks to the cloud layer; wherein the second type of tasks are tasks characterized by centralized planning by the cloud brain and distributed implementation by the edge; The edge layer is also used to provide lightweight computing, analysis, processing and decision-making; The cloud layer is used to receive the processing request of the first type of task sent by the edge layer, process the first type of task, obtain the processing result of the first type of task, and return the processing result of the first type of task to the edge layer; The cloud layer is further used to send the second type of tasks to the edge layer and receive the processing results of the second type of tasks reported by the edge layer; The cloud layer includes a mode management unit, a resource scheduling unit, an alarm unit, a mirror unit, various types of databases, an operation status analysis unit, a simulation and emulation unit, an artificial intelligence unit, a visualization unit, an intelligent operation and maintenance unit, and a security management unit; The mode management unit is used to configure and manage the mode according to the plan and objectives of the operation of the system to be controlled, and to issue and allocate all tasks under the configured mode; The resource scheduling unit is used to dynamically and flexibly schedule and use computing, storage and network resources according to various task requirements; The alarm unit is used to push sound, light or SMS messages according to the set strategy and priority of the device status of the abnormal device according to the analysis results of the operating status of each device in the system to be controlled; The mirror unit is used to package various standard or customized systems, services, and software resources into file resources through containerization and virtualization, so that the intelligent operation and maintenance unit can retrieve, deploy, and perform intelligent installation and configuration; The various types of databases include one or more of an artificial intelligence model library, a visualization component library, a function library, a document library, and a configuration parameter library; The operation status analysis unit is used to monitor and analyze the operation parameters of each device, software service and device of the controlled system; The simulation and emulation unit is used to simulate and emulate the execution process of the cloud layer; The artificial intelligence unit is used for one or more of fault prediction, fault analysis, health diagnosis, and automatic beam adjustment; The visualization unit is used to display the filtered and analyzed data based on the visualization component library; The intelligent operation and maintenance unit is used to perform intelligent operation and maintenance management on the hardware devices and virtual machines involved in the cloud-edge-device architecture; The security management unit is used to provide security capabilities for computing, storage and network resources within the cloud layer.

2. The distributed intelligent control platform according to claim 1, characterized in that: The system to be controlled includes an accelerator, and the various devices of the accelerator connected to the end layer include one or more of a charged particle generating device, a vacuum acceleration chamber, a beam transport system, a beam monitoring and diagnostic device, an electromagnetic field stabilization control device, and a power supply and operating device.

3. The distributed intelligent control platform according to claim 1, characterized in that: The cloud layer includes cloud brain, cloud storage, cloud computing and cloud security; The cloud brain is used to manage and coordinate plans, modes, models, computing, storage and network resources within the cloud layer; The cloud storage is used to store data and provide data support for cloud computing; The cloud computing is used to process and analyze data; The cloud security is used to provide security capabilities for computing, storage and network resources within the cloud layer.

4. The distributed intelligent control platform according to claim 1, characterized in that: The cloud layer also includes a user interaction unit for receiving plans and goals for operation of the system to be controlled submitted by a user.

5. The distributed intelligent control platform according to claim 1, characterized in that: The edge layer includes several edge nodes, a shadow implementation unit, a message management unit, a real-time calculation unit, a snapshot storage unit, and a regional decision unit; The shadow implementation unit is used to copy the service mappings of the cloud layer to the edge node; The message management unit is used to provide message communication and collaboration mechanisms for different modules and services; The real-time calculation unit is used to generate real-time control instructions based on the data of multiple functional modules and various devices of the system to be controlled; The snapshot storage unit is used to temporarily store the snapshot generated by the edge layer; The regional decision unit is used to perform regional decision analysis and processing on the second type of tasks sent by the cloud layer.

6. The distributed intelligent control platform according to claim 1, characterized in that: The edge layer is used to provide the underlying hardware of multiple functional modules for modularly controlling each device of the controlled system, and is a board for realizing configurable control functions.

7. The distributed intelligent control platform according to claim 6, characterized in that: The board includes a baseboard, a core board, and a functional sub-board; The core board is pluggably connected to the base board; The functional sub-board is pluggably connected to the base board.

8. The distributed intelligent control platform according to claim 7, characterized in that: The core board is implemented based on a field programmable gate array (FPGA).

9. The distributed intelligent control platform according to claim 1, characterized in that: The control platform includes a data network and a real-time network; The data network is used for transmitting various types of data of the cloud layer, the edge layer and the terminal layer; The real-time network is used for timing triggering, time synchronization and transmission of real-time data; The data network and the real-time network are constructed by wired or wireless means, with different levels of time accuracy, providing time references from ns, us, ms to s, and completing intelligent control according to strict logical timing.

Images