CN113742673A - Cloud edge collaborative management and control integrated platform based on federal learning - Google Patents

Cloud edge collaborative management and control integrated platform based on federal learning Download PDF

Info

Publication number
CN113742673A
CN113742673A CN202111045212.8A CN202111045212A CN113742673A CN 113742673 A CN113742673 A CN 113742673A CN 202111045212 A CN202111045212 A CN 202111045212A CN 113742673 A CN113742673 A CN 113742673A
Authority
CN
China
Prior art keywords
integrated platform
cloud
management
module
federal learning
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202111045212.8A
Other languages
Chinese (zh)
Other versions
CN113742673B (en
Inventor
石硕
刘秉哲
齐学海
刘彦华
赵强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Daohe Wulian Technology Co ltd
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CN202111045212.8A priority Critical patent/CN113742673B/en
Publication of CN113742673A publication Critical patent/CN113742673A/en
Application granted granted Critical
Publication of CN113742673B publication Critical patent/CN113742673B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/30Authentication, i.e. establishing the identity or authorisation of security principals
    • G06F21/31User authentication
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60Protecting data
    • G06F21/602Providing cryptographic facilities or services
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60Protecting data
    • G06F21/62Protecting access to data via a platform, e.g. using keys or access control rules
    • G06F21/6218Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60Protecting data
    • G06F21/64Protecting data integrity, e.g. using checksums, certificates or signatures

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Health & Medical Sciences (AREA)
  • Bioethics (AREA)
  • General Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Information Transfer Between Computers (AREA)
  • Stored Programmes (AREA)

Abstract

A cloud side cooperative management and control integrated platform based on federal learning comprises an integrated platform, wherein the integrated platform is designed by a B/S architecture and is used for executing functions of model training, equipment management, picture configuration, logic configuration, compiling and downloading and simulating simulation; the cloud fort machine is connected with the integrated platform through an intranet and used for executing functions of identity verification, account management, authorization control and safety audit; the cloud fort machine is connected with a plurality of small DCS all-in-one machines through a public network IP, and the federal learning function is completed through the steps of public key distribution, encrypted transmission, result gathering and model updating.

Description

Cloud edge collaborative management and control integrated platform based on federal learning
The technical field is as follows:
the invention relates to a cloud-side cooperative management and control integrated platform based on federal learning.
Background art:
with the rapid development of artificial intelligence technology, the application of the deep learning network model to equipment fault detection has great advantages, valuable information can be mined from mass data, and the running state of equipment can be judged according to the running data of the equipment.
The federated learning method can perform joint training of the model among the devices with the same characteristics, expand the data scale and enhance the generalization capability of the model in the migration application among the devices; however, the existing deep neural network model has complex structure, high training and calculation cost, is difficult to deploy in industrial edge equipment, has limited capability of extracting data characteristics outside a time characteristic domain, and gradually presents the characteristics of massive isomerism, complex processing, high calculation frequency and the like under the background of large-scale popularization of industrial internet-of-things terminal equipment.
At present, a commonly used cloud computing architecture usually directly uploads perception layer data to a cloud platform for centralized processing and application, and as the total amount of data is large, a phenomenon of large transmission delay exists in an actual transmission process, which easily causes problems of untimely service response and large platform-side network resource burden.
The invention content is as follows:
the embodiment of the invention provides a cloud-side cooperative management and control integrated platform based on federal learning, which is reasonable in structural design, can accurately and quickly transmit or upload data through the mutual cooperation of a plurality of platforms and functional parts, can keep higher working efficiency even if the total amount of the data is larger, reduces transmission delay and resource burden, enables services or operations to respond in time according to instructions of users, reduces the probability of risk occurrence, ensures the safety in the data transmission process, prevents deviation, and solves the problems in the prior art.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a cloud side cooperative management and control integrated platform based on federal learning comprises an integrated platform, wherein the integrated platform is designed by a B/S architecture and is used for executing functions of model training, equipment management, picture configuration, logic configuration, compiling and downloading and simulating simulation; the cloud fort machine is connected with the integrated platform through an intranet and used for executing functions of identity verification, account management, authorization control and safety audit; the cloud fort machine is connected with a plurality of small DCS all-in-one machines through a public network IP, and the federal learning function is completed through the steps of public key distribution, encrypted transmission, result gathering and model updating.
The integrated platform is deployed in the server through the EXSI, and the virtual DPUs in the integrated platform are deployed in the server in a hardware direct connection mode through the EXSI alone.
Be equipped with management and control equipment on the cloud fort machine, thereby management and control equipment is used for accomplishing equipment activation and gets through the connection of integration platform, cloud fort machine and miniaturized DCS all-in-one.
And realizing a logic configuration function by using Java programming language in the integrated platform to obtain a logic configuration file, and downloading the logic configuration file to a JVM virtual machine in a virtual DPU in the integrated platform after compiling the logic configuration file so as to realize cross-platform operation of logic configuration.
And an upper-layer big data platform is connected to the integrated platform to realize remote communication of data.
And a digital twin simulation system is also connected to the integrated platform to realize analog simulation and high-efficiency data transmission.
The integrated platform is internally provided with a model training module, an equipment management module, a picture configuration module, a logic configuration module, a compiling and downloading module, a user management module, a monitoring alarm module, an operation monitoring module, a data management module and an analog simulation module.
An identity authentication module, an account management module, an authorization control module and a security audit module are arranged in the cloud bastion.
By adopting the structure, the functions of model training, equipment management, picture configuration, logic configuration, compiling and downloading and simulation are executed through the integrated platform so as to finish the transmission and processing of data; the cloud bastion machine is used for executing the functions of identity verification, account management, authorization control and security audit so as to establish man-machine interaction with a user and input a control instruction; completing equipment activation through control equipment to open the connection of the integrated platform, the cloud fort machine and the miniaturized DCS integrated machine; the logic configuration function is realized by using Java programming language in the integrated platform to obtain the logic configuration file, and the integrated platform has the advantages of accuracy, practicability, safety and reliability.
Description of the drawings:
FIG. 1 is a schematic structural diagram of the present invention.
FIG. 2 is a flow chart of the training of the learning model of the present invention.
Fig. 3 is a functional module diagram of the integrated platform of the present invention.
Fig. 4 is a functional module schematic diagram of the cloud fort machine of the invention.
The specific implementation mode is as follows:
in order to clearly explain the technical features of the present invention, the following detailed description of the present invention is provided with reference to the accompanying drawings.
As shown in fig. 1 to 4, a federated learning-based cloud-side collaborative management and control integrated platform includes an integrated platform, which is designed by a B/S architecture and is used for executing functions of model training, device management, screen configuration, logic configuration, compiling and downloading, and simulation; the cloud fort machine is connected with the integrated platform through an intranet and used for executing functions of identity verification, account management, authorization control and safety audit; the cloud fort machine is connected with a plurality of small DCS all-in-one machines through a public network IP, and the federal learning function is completed through the steps of public key distribution, encrypted transmission, result gathering and model updating.
The integrated platform is deployed in the server through the EXSI, and the virtual DPUs in the integrated platform are deployed in the server in a hardware direct connection mode through the EXSI alone.
Be equipped with management and control equipment on the cloud fort machine, thereby management and control equipment is used for accomplishing equipment activation and gets through the connection of integration platform, cloud fort machine and miniaturized DCS all-in-one.
And realizing a logic configuration function by using Java programming language in the integrated platform to obtain a logic configuration file, and downloading the logic configuration file to a JVM virtual machine in a virtual DPU in the integrated platform after compiling the logic configuration file so as to realize cross-platform of logic configuration.
And an upper-layer big data platform is connected to the integrated platform to realize remote communication of data.
And a digital twin simulation system is also connected to the integrated platform to realize analog simulation and high-efficiency data transmission.
The integrated platform is internally provided with a model training module, an equipment management module, a picture configuration module, a logic configuration module, a compiling and downloading module, a user management module, a monitoring alarm module, an operation monitoring module, a data management module and an analog simulation module.
An identity authentication module, an account management module, an authorization control module and a security audit module are arranged in the cloud bastion.
The working principle of the cloud-side cooperative management and control integrated platform based on federal learning in the embodiment of the invention is as follows: through the mutually supporting effect of a plurality of platforms and functional unit, can be accurate quick carry out the transmission or the upload of data, even the total amount of data is great, also can keep higher work efficiency, transmission delay and resource burden have been reduced, make business or operation in time respond according to user's instruction, the probability that the risk takes place has been reduced simultaneously, guarantee the security in the data transmission process, prevent the production of deviation, the application scene is comparatively extensive, the popularization degree of difficulty is little, can promote at industrial production's in-process fast.
In the integral scheme, the integrated platform and the cloud fort machine are connected through an internal network, the cloud fort machine is connected with a plurality of small DCS all-in-one machines through a public network IP, and the federal learning function is completed through the steps of public key distribution, encrypted transmission, result gathering and model updating, so that accurate and rapid transmission of a large amount of data is realized by means of the structure; meanwhile, the whole structure is developed in a low-coupling component library development mode, each functional component can be independently expanded and upgraded, the use of the whole function is not influenced, and the stability of platform application is improved.
Preferably, the integrated platform is designed by adopting a B/S architecture, the integrated platform is deployed in the server through EXSI, and the virtual DPU in the integrated platform is deployed in the server in a hardware direct connection mode through the EXSI alone; the EXSI is specially designed for running a virtual machine, reducing configuration requirements to the maximum extent and simplifying deployment; the client can complete the whole process of installing the running virtual machine only in a few minutes, particularly when downloading and installing the pre-configured virtual equipment; the deployment, operation and maintenance of multiple platforms and multiple virtual DPUs can be easily realized through the EXSI.
Preferably, the cloud fort machine is provided with the control equipment, so that the equipment activation is completed, the connection of the integrated platform, the cloud fort machine and the miniaturized DCS all-in-one machine is opened, and a user can conveniently trigger the whole structure to enter a working state.
After the network connection, the cloud fort machine waits for the further activation authentication of the integrated platform, the equipment activation code of each small DCS all-in-one machine can be input into the equipment management module of the integrated platform, the authentication of the small DCS all-in-one machine on the integrated platform side is completed, the management function of the integrated platform on the equipment is activated, and the integrated platform can activate the small DCS all-in-one machines according to the operation.
Preferably, a Java programming language is used in the integrated platform to realize a logic configuration function, so that a logic configuration file is obtained, and the logic configuration file is compiled and then downloaded to a JVM virtual machine in a virtual DPU in the integrated platform, so that cross-platform logic configuration is realized; the generated file of the logic configuration is a java file. And finally a class file after being compiled.
An upper-layer big data platform can be connected to the integrated platform to realize remote communication of data, and real-time data and non-real-time data are stored in the cloud by Kafka and Flume under normal conditions; a large amount of operation data in the actual industrial process can be compressed and uploaded by a specific compression algorithm, and a large data platform function API can be called on the integrated platform function.
As shown in fig. 2, taking two devices as an example, a collaborator C distributes a public key to a device a and a device B for encrypting data to be exchanged during a training process; the device A and the device B interact in an encrypted form to calculate an intermediate result of the gradient; and the equipment A and the equipment B respectively calculate based on the encrypted gradient values, meanwhile, the equipment B calculates loss according to the label data of the equipment B, summarizes the result to the collaborator C, calculates the total gradient value through the summarized result and decrypts the total gradient value, the collaborator C respectively returns the decrypted gradient to the equipment A and the equipment B, and the equipment A and the equipment B respectively update the parameters of the respective models according to the gradient.
Iterating the steps until the loss function is converged, thereby completing the whole model training process; in the sample alignment and model training process, the data of the device A and the data of the device B are both kept locally, and data privacy disclosure cannot be caused by data interaction in training. Thus, both parties are enabled to collaboratively train the model with the help of federal learning.
The cloud configuration in the integrated platform can be divided into a cloud picture configuration and a cloud logic configuration, specifically, the cloud picture configuration is realized by means of 2D and 3D model component libraries and by means of drag-and-drop programming, and the file format finally generated by the picture configuration is XML. Wherein the 3D model component library can be realized by three.
The cloud logic configuration is realized by using Java programming language to realize building block type programming of a web end, and a user finishes the construction of configuration logic by dragging a code block generated by Java.
Specifically, the configuration of the cloud screen, the configuration of the cloud logic, and the configured monitoring screen need to be completed by means of a set of component databases. Namely, the data point location channel information required by the picture configuration logic configuration is preset in the database, and the appropriate configuration is selected to complete the reading and writing of the corresponding data and the establishment of the control logic.
And for data point positions related in the picture configuration and the logic configuration, the data point positions can be synchronized to a MongoDB database in the cloud management and control integrated platform at the edge side through modes of OPC, MQTT, HTTP and the like. And the cloud picture configuration, the cloud logic configuration and the field DPU equipment build a data bridge by depending on the MongoDB.
For the cloud side communication in the embodiment of the invention, firstly, the edge side equipment is powered on and activated, then the edge side equipment is registered and survived to the cloud bastion machine in a TCPIP mode, the edge side equipment is further confirmed and activated at the integrated platform end, then the cloud bastion machine can be used as a central node, and the two sides of the cloud side are used as child nodes to establish a virtual local area network.
For the connection digital twin system in the embodiment of the present invention, firstly, the implementation manner of the digital twin system includes, but is not limited to, modeica. Modelica is an existing physical system modeling language that supports the development of efficient model libraries and model reuse. The construction of the simulation system can be completed by using Modelica. The communication principle of the virtual DPU is basically consistent with that of a real DPU. Under a real application scene, the DPU is connected with various input and output clamping pieces through a CAN bus, and the various input and output clamping pieces support the collection of various industrial signals. In practical applications, the digital twin system needs to develop a data input and output block including but not limited to CAN communication. In the actual connection, the digital twin system and the virtual DPU communicate with each other through a virtual CAN bus. The data input and output block in the digital twin system can be configured in the integrated platform consistent with the configuration of the real equipment. The method can be expanded between the virtual DPUs of special projects through Modbus and OPC.
In summary, the cloud-side cooperative management and control integrated platform based on federal learning in the embodiments of the present invention can accurately and quickly transmit or upload data through the interaction of the multiple platforms and the functional components, so that even if the total amount of data is large, high work efficiency can be maintained, transmission delay and resource load are reduced, services or operations can timely respond according to the instructions of users, meanwhile, the probability of risk occurrence is reduced, the security in the data transmission process is ensured, the generation of deviations is prevented, the application scenarios are wide, the popularization difficulty is low, and the cloud-side cooperative management and control integrated platform can be quickly popularized in the industrial production process.
The above-described embodiments should not be construed as limiting the scope of the invention, and any alternative modifications or alterations to the embodiments of the present invention will be apparent to those skilled in the art.
The present invention is not described in detail, but is known to those skilled in the art.

Claims (8)

1. The utility model provides a management and control integration platform is in coordination with cloud limit based on federal study which characterized in that: the system comprises an integrated platform, wherein the integrated platform is designed by adopting a B/S architecture, and is used for executing the functions of model training, equipment management, picture configuration, logic configuration, compiling and downloading and simulating simulation; the cloud fort machine is connected with the integrated platform through an intranet and used for executing functions of identity verification, account management, authorization control and safety audit; the cloud fort machine is connected with a plurality of small DCS all-in-one machines through a public network IP, and the federal learning function is completed through the steps of public key distribution, encrypted transmission, result gathering and model updating.
2. The cloud side collaborative management and control integrated platform based on federal learning according to claim 1, characterized in that: the integrated platform is deployed in the server through the EXSI, and the virtual DPUs in the integrated platform are deployed in the server in a hardware direct connection mode through the EXSI alone.
3. The cloud side collaborative management and control integrated platform based on federal learning according to claim 1, characterized in that: be equipped with management and control equipment on the cloud fort machine, thereby management and control equipment is used for accomplishing equipment activation and gets through the connection of integration platform, cloud fort machine and miniaturized DCS all-in-one.
4. The cloud side collaborative management and control integrated platform based on federal learning according to claim 1, characterized in that: and realizing a logic configuration function by using Java programming language in the integrated platform to obtain a logic configuration file, and downloading the logic configuration file to a JVM virtual machine in a virtual DPU in the integrated platform after compiling the logic configuration file so as to realize cross-platform of logic configuration.
5. The cloud side collaborative management and control integrated platform based on federal learning according to claim 1, characterized in that: and an upper-layer big data platform is connected to the integrated platform to realize remote communication of data.
6. The cloud side collaborative management and control integrated platform based on federal learning according to claim 1, characterized in that: and a digital twin simulation system is also connected to the integrated platform to realize analog simulation and high-efficiency data transmission.
7. The cloud side collaborative management and control integrated platform based on federal learning according to claim 1, characterized in that: the integrated platform is internally provided with a model training module, an equipment management module, a picture configuration module, a logic configuration module, a compiling and downloading module, a user management module, a monitoring alarm module, an operation monitoring module, a data management module and an analog simulation module.
8. The cloud side collaborative management and control integrated platform based on federal learning according to claim 1, characterized in that: an identity authentication module, an account management module, an authorization control module and a security audit module are arranged in the cloud bastion.
CN202111045212.8A 2021-09-07 2021-09-07 Cloud edge cooperative control integrated platform based on federal learning Active CN113742673B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111045212.8A CN113742673B (en) 2021-09-07 2021-09-07 Cloud edge cooperative control integrated platform based on federal learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111045212.8A CN113742673B (en) 2021-09-07 2021-09-07 Cloud edge cooperative control integrated platform based on federal learning

Publications (2)

Publication Number Publication Date
CN113742673A true CN113742673A (en) 2021-12-03
CN113742673B CN113742673B (en) 2024-05-10

Family

ID=78736701

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111045212.8A Active CN113742673B (en) 2021-09-07 2021-09-07 Cloud edge cooperative control integrated platform based on federal learning

Country Status (1)

Country Link
CN (1) CN113742673B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111857065A (en) * 2020-06-08 2020-10-30 北京邮电大学 Intelligent production system and method based on edge calculation and digital twinning
CN112003924A (en) * 2020-08-20 2020-11-27 浪潮云信息技术股份公司 Industrial internet-oriented edge cloud platform building method and system
CN112232528A (en) * 2020-12-15 2021-01-15 之江实验室 Method and device for training federated learning model and federated learning system
WO2021092980A1 (en) * 2019-11-14 2021-05-20 深圳前海微众银行股份有限公司 Longitudinal federated learning optimization method, apparatus and device, and storage medium
CN113112029A (en) * 2021-04-22 2021-07-13 中国科学院计算技术研究所 Federal learning system and method applied to heterogeneous computing equipment

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021092980A1 (en) * 2019-11-14 2021-05-20 深圳前海微众银行股份有限公司 Longitudinal federated learning optimization method, apparatus and device, and storage medium
CN111857065A (en) * 2020-06-08 2020-10-30 北京邮电大学 Intelligent production system and method based on edge calculation and digital twinning
CN112003924A (en) * 2020-08-20 2020-11-27 浪潮云信息技术股份公司 Industrial internet-oriented edge cloud platform building method and system
CN112232528A (en) * 2020-12-15 2021-01-15 之江实验室 Method and device for training federated learning model and federated learning system
CN113112029A (en) * 2021-04-22 2021-07-13 中国科学院计算技术研究所 Federal learning system and method applied to heterogeneous computing equipment

Also Published As

Publication number Publication date
CN113742673B (en) 2024-05-10

Similar Documents

Publication Publication Date Title
Mashaly Connecting the twins: A review on digital twin technology & its networking requirements
CN105843182B (en) A kind of power scheduling accident prediction system and method based on OMS
Neema et al. Model-based integration platform for FMI co-simulation and heterogeneous simulations of cyber-physical systems
CN109460593B (en) Block chain virtual simulation system
CN104035392A (en) Big data in process control systems
CN103645908B (en) Full life circle development achievement system of intemetware
CN112417037A (en) Block chain construction method for distributed identity authentication in industrial field
Shkarupylo et al. Development of stratified approach to software defined networks simulation
CN105049420A (en) Security protocol formal verification method using expanded UML model as framework
CN115102827A (en) Digital product real-time monitoring general Internet platform for small and medium-sized manufacturing industry
CN109478055A (en) Using intelligent node for monitoring industrial process in general, intelligence system
Kosek et al. Evaluation of smart grid control strategies in co-simulation—Integration of IPSYS and mosaik
Mazur et al. Analysis of possible SDN use in the rapid prototyping process as part of the Industry 4.0
CN115439249B (en) Method, device and system for realizing business service of cross-region block chain
CN109274756A (en) A kind of intelligent environment protection monitoring management system
CN115099068B (en) Oil reservoir numerical simulation system based on metauniverse
CN106878333B (en) A kind of third party authentication method, device and application system server
TWI651667B (en) Interface system of virtual and physical integrated network with resources topology
Mathioudakis et al. Towards generic SCADA simulators: A survey of existing multi-purpose co-simulation platforms, best practices and use-cases
Rellermeyer et al. Building, deploying, and monitoring distributed applications with eclipse and r-osgi
CN112486666A (en) Model-driven reference architecture method and platform
CN113742673A (en) Cloud edge collaborative management and control integrated platform based on federal learning
CN105468431B (en) Multi-model real-time emulation system
Sztipanovits et al. Model-based integration technology for next generation electric grid simulations
CN105790943A (en) License pre-authentication method, device and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right

Effective date of registration: 20240511

Address after: Building 9-117-132, Zhongrun Century City, No. 13777 Jingshi Road, Lixia District, Jinan City, Shandong Province, 250000

Patentee after: Shandong Daohe Wulian Technology Co.,Ltd.

Country or region after: China

Address before: # 8-1903, Zone 8, Luneng Lingxiu City, Shizhong District, Jinan City, Shandong Province, 250000

Patentee before: Shi Shuo

Country or region before: China