WO2022252717A1 - 一种实现电力数据处理的同异构混合多核芯片架构 - Google Patents

一种实现电力数据处理的同异构混合多核芯片架构 Download PDF

Info

Publication number
WO2022252717A1
WO2022252717A1 PCT/CN2022/078178 CN2022078178W WO2022252717A1 WO 2022252717 A1 WO2022252717 A1 WO 2022252717A1 CN 2022078178 W CN2022078178 W CN 2022078178W WO 2022252717 A1 WO2022252717 A1 WO 2022252717A1
Authority
WO
WIPO (PCT)
Prior art keywords
power
processing
module
data processing
real
Prior art date
Application number
PCT/CN2022/078178
Other languages
English (en)
French (fr)
Inventor
李鹏
李立浧
习伟
黄凯
姚浩
陈军健
蒋小文
陶伟
邓清唐
于杨
Original Assignee
南方电网数字电网研究院有限公司
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 南方电网数字电网研究院有限公司 filed Critical 南方电网数字电网研究院有限公司
Publication of WO2022252717A1 publication Critical patent/WO2022252717A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00—Digital computers in general; Data processing equipment in general
    • G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
    • G06F15/163—Interprocessor communication
    • G06F15/167—Interprocessor communication using a common memory, e.g. mailbox
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00—Digital computers in general; Data processing equipment in general
    • G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
    • G06F15/163—Interprocessor communication
    • G06F15/173—Interprocessor communication using an interconnection network, e.g. matrix, shuffle, pyramid, star, snowflake
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Definitions

  • the present application relates to the technical field of power grids, in particular to a homogeneous hybrid multi-core chip architecture for realizing power data processing and a method for power data processing.
  • edge computing has become a new node that carries grid services and data.
  • a large amount of data is aggregated and fused at the edge, and multi-services are coordinated and processed locally at the edge.
  • the digital grid edge side generally does not form new physical nodes, and is integrated with existing devices.
  • Highly concurrent applications and diverse data processing requirements require edge side core chips to have high performance and high security. , low power consumption and other features to meet the key needs of real-time response, data security, and intelligent processing.
  • Edge-side computing chips generally adopt the SMP homogeneous multi-core architecture, lacking heterogeneous computing units oriented to the computing needs of power grid business application scenarios, chip performance and computing modes are difficult to meet the growing edge computing performance requirements, and lacking edge-side multi-service data fusion It is difficult to adapt to the development trend of power grid digitization due to security considerations.
  • a homogeneous hybrid multi-core chip architecture for power data processing includes a homogeneous main control unit and a heterogeneous data processing unit; wherein:
  • the heterogeneous data processing unit includes data processing sub-units required by different power service requirements; the isomorphic main control unit acquires power data and generates the power service demand correspondence by responding to a service processing request carrying power service requirements a control instruction; the target data processing subunit in the heterogeneous data processing unit processes the power data according to the processing request by receiving the processing request sent by the homogeneous main control unit.
  • the target data processing subunit includes at least one of an intelligent operation processing subunit, a security processing subunit, a real-time processing subunit, and a power algorithm processing subunit;
  • the isomorphic main control unit is a multi-core Isomorphic main control unit.
  • the security processing subunit is used for isolating and encrypting the power data for security processing;
  • the real-time processing subunit is used for real-time processing of the power data required by the power service with high real-time requirements;
  • the power algorithm processing subunit is used to determine the corresponding power algorithm to process the power data according to the processing performance requirements or data processing mode requirements;
  • the intelligent operation processing subunit is used to process the power data according to the neural network algorithm deal with.
  • the isomorphic main control unit adopts a high-performance core
  • the safety processing subunit adopts a safety core
  • the real-time processing subunit adopts a real-time core
  • the power algorithm processing subunit adopts a dedicated processing device
  • the intelligent operation processing sub-unit adopts an intelligent operation processor.
  • the power algorithm processing subunit communicates with the real-time processing subunit through a private bus, and the real-time processing subunit has independent on-chip storage and The power algorithm processing subunits are connected.
  • a method for processing power data based on a homogeneous hybrid multi-core chip comprising:
  • the heterogeneous data processing module includes data processing sub-modules required by different power service requirements
  • the target data processing sub-module includes a power algorithm processing sub-module and a real-time processing sub-module;
  • the target data processing sub-module in the structural data processing module including:
  • the controlling the target data processing sub-module to process the power data according to the control instruction includes:
  • the target data processing sub-module includes an intelligent operation processing sub-module; when the power business demand is non-real-time demand, the control instruction is sent to the heterogeneous data processing module
  • the target data processing submodules in include:
  • the controlling the target data processing sub-module to process the power data according to the control instruction includes:
  • Machine learning is performed on the power data according to the neural network algorithm to obtain implicit data representing the safety of the power grid.
  • the system architecture memory stores a computer program, and the processor implements the following steps when executing the computer program:
  • the heterogeneous data processing module includes data processing sub-modules required by different power service requirements
  • the system architecture readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:
  • the heterogeneous data processing module includes data processing sub-modules required by different power service requirements
  • the above homogeneous hybrid multi-core chip architecture that realizes power data processing obtains power data and generates control instructions corresponding to power business needs through the isomorphic main control unit in response to business processing requests that carry power business needs, including different power business needs.
  • the required heterogeneous data processing units determine the target data processing sub-unit for processing power business requirements; the target data processing sub-unit processes the power data according to the processing request by receiving the processing request sent by the isomorphic main control unit, that is, using the same
  • the heterogeneous hybrid multi-chip realizes multi-core concurrent processing and multiple heterogeneous processing of power data, ensuring power data processing efficiency and power grid data security in different application scenarios.
  • FIG. 1 is a schematic diagram of a homogeneous hybrid multi-core chip architecture for realizing power data processing in an embodiment
  • FIG. 2 is a schematic diagram of a heterogeneous hybrid multi-core chip architecture for realizing power data processing in another embodiment
  • FIG. 3 is a schematic diagram of a heterogeneous hybrid multi-core chip architecture for implementing power data processing in another embodiment
  • FIG. 4 is a schematic flow diagram of a power data processing method based on a heterogeneous hybrid multi-core chip architecture in an embodiment
  • FIG. 5 is an application scenario diagram of a power data processing method based on a homogeneous hybrid multi-core chip architecture in an embodiment
  • Fig. 6 is an internal structure diagram of the electric device in one embodiment.
  • a homogeneous hybrid multi-core chip architecture for power data processing is provided.
  • the application of the system to a terminal is used as an example for illustration. It can be understood that the system also Can be applied to the server.
  • the system includes a homogeneous main control unit and a heterogeneous data processing unit, wherein:
  • the isomorphic main control unit obtains power data and generates control instructions corresponding to power business requirements by responding to business processing requests carrying power business requirements.
  • the homogeneous main control unit and the heterogeneous data processing unit are integrated on the homogeneous hybrid multi-core chip;
  • Design heterogeneous means that multiple data processing units adopt different types of computing units, which can meet different processing requirements of different power data.
  • the homogeneous main control unit and the heterogeneous data processing unit are based on the communication architecture for data interaction; the communication architecture is composed of an on-chip interconnection network and private interconnection and storage, where the on-chip interconnection network is applied between the units of the homogeneous hybrid multi-core chip
  • the private interconnection exists only for communication between specific data processing subunits (for example, real-time processing subunit and power algorithm subunit).
  • Power business requirements are business requirements at the edge of the power grid (for example, in SCADA systems, EMS, substation automation systems, converter station computer monitoring systems, power plant computer monitoring systems, distribution automation systems, microcomputer relay protection and safety automatic devices) Different business requirements), business processing requests include real-time processing of power data, prediction and analysis of power grid failures, isolation or encryption of different types of power data, and power grid frequency control.
  • the main control core of the homogeneous main control unit will control the data collection, and at the same time send control instructions to the real-time processing sub-unit in the heterogeneous data processing unit to control the real-time
  • the processing subunit completes the calculation of the corresponding protection parameters; when communication is required, it sends a control command to the security processing subunit, controls the security processing subunit to encrypt the protection parameter data, and communicates the obtained encrypted data.
  • the heterogeneous data processing unit includes data processing subunits required by different power service requirements (such as data processing subunit 1, data processing subunit 2,...data processing subunit n); the target in the heterogeneous data processing unit
  • the data processing subunit processes the power data according to the processing request by receiving the processing request sent by the isomorphic main control unit.
  • the heterogeneous data processing unit includes an intelligent operation processing sub-unit, a security processing sub-unit, a real-time processing sub-unit, and a power algorithm processing sub-unit.
  • the power algorithm processing subunit refers to a kind of algorithm processing applied in power scenarios proposed according to the characteristics or needs of the power grid, including pre-data processing algorithms, electrical parameter calculation algorithms, data management algorithms, and network communication algorithms, etc.;
  • the pre-data processing and parameter calculation have clear targeting, that is, the data, parameters, data management mode and operation requirements required by the power can be targeted;
  • the network communication algorithm has the specification constraints of the power standard, and is suitable for special processing device to achieve acceleration.
  • the homogeneous main control unit obtains power data and generates control instructions corresponding to power business needs by responding to business processing requests carrying power business needs, and the heterogeneous data processing unit
  • the target data processing sub-unit processes the power data by responding to the control instructions sent by the isomorphic main control unit, that is, the isomorphic main control unit is used to implement task scheduling among multiple heterogeneous data processing units, and process control instructions and Control instruction execution process, that is, a heterogeneous hybrid multi-core chip based on multiple unit configurations, improves the heterogeneous computing capability of the chip, and ensures that the chip can meet the requirements of chip-level high performance, low latency and Intelligent and diverse computing requirements improve the efficiency and security of power data processing at the edge of the grid.
  • a homogeneous hybrid multi-core chip architecture for power data processing is provided.
  • the application of the system to a terminal is used as an example for illustration. Can be applied to the server.
  • a homogeneous main control unit and a heterogeneous data processing unit are included; wherein:
  • the heterogeneous data processing unit includes intelligent operation processing sub-units, security processing sub-units, real-time processing sub-units and power algorithm processing sub-units; the homogeneous main control unit obtains power data and Generate control instructions corresponding to power business requirements; the target data processing subunit in the heterogeneous data processing unit processes the power data by responding to the control instructions sent by the homogeneous main control unit.
  • the security processing subunit is used for isolating and encrypting power data for safe processing;
  • the real-time processing subunit is used for real-time processing of power data required by high real-time power business requirements;
  • the power algorithm processing subunit is used for processing according to processing performance requirements or
  • the data processing method requires to determine the corresponding power algorithm to process the power data;
  • the intelligent operation processing subunit is used to process the power data according to the neural network algorithm, and obtain the implicit data representing the safety of the power grid.
  • the isomorphic main control unit is a multi-core isomorphic main control unit.
  • the isomorphic main control unit adopts a high-performance core and a multi-core isomorphic architecture;
  • the safety processing sub-unit adopts a safe core; real-time The processing sub-unit adopts real-time core;
  • the power algorithm processing sub-unit adopts a special processor;
  • the intelligent operation processing sub-unit adopts an intelligent operation processor.
  • Each processor core performs separate calculations for different services without interfering with each other. When a processor core fails, it will not cause failure to other processor cores, which improves the processing efficiency of power data and ensures the power grid. safety.
  • the communication between the isomorphic main control unit and the intelligent operation processing sub-unit, the security processing sub-unit, the real-time processing sub-unit and the power algorithm processing sub-unit is carried out through the Internet on chip, that is, the Internet on chip that can be scheduled and preempted is used to connect the system. All units, including high-performance cores, safety cores, real-time cores, and intelligent computing processors, improve the efficiency of power data processing and the reasonable utilization of communication resources.
  • the on-chip interconnection network includes mailbox (mailbox) and interrupt (intc) to ensure information communication between high-performance cores, safety cores, real-time cores, and intelligent computing processors.
  • Each core can trigger an interrupt source of the other party.
  • Mailbox is a peripheral module used for inter-core communication on a multi-core chip, It can be called by multiple cores, set multiple channels, and can receive interrupt signals.
  • the homogeneous hybrid multi-core chip architecture for power data processing also includes private interconnection and storage. Private interconnection and storage exist in the real-time subsystem composed of the real-time processing subunit and the power algorithm processing subunit.
  • the power algorithm processing subsystem is connected to the real-time core of the real-time computing module through a private bus to ensure that power service processing requests can be quickly responded .
  • the main control unit adopts a CPU based on a domestic instruction set, which can realize domestic independent controllability.
  • the power algorithm processing subunit can realize highly customized algorithm types, according to various power application scenarios , such as fault diagnosis, status monitoring, load identification, power outage identification, load control and other commonly used algorithms, design and integrate different computing parts to meet the corresponding demand functions, and truly realize power-specific.
  • the security processing subunit can integrate multiple data encryption algorithms and security protection algorithms to achieve security defense and attack immunity on the edge side.
  • the intelligent computing module uses a low-power, programmable neural network processor, which is suitable for intelligent analysis and processing of power terminals.
  • a homogeneous hybrid multi-core chip architecture for realizing power data processing in one embodiment, wherein the communication architecture is composed of an on-chip interconnection network and private interconnection and storage, and the homogeneous main control unit It is a multi-core isomorphic main control unit, the safety processing sub-unit adopts a safety core, the real-time processing sub-unit adopts a real-time core, the power algorithm processing sub-unit adopts a special processor, and the intelligent operation processing sub-unit adopts an intelligent operation processor; the isomorphic main control unit Communication with the intelligent computing processing subunit, security processing subunit, real-time processing subunit and power algorithm processing subunit is based on the on-chip interconnection network, that is, there is a private interconnection between the real-time processing subunit and the power algorithm processing subunit, and real-time processing The sub-unit and the power algorithm processing sub-unit communicate based on private interconnection; the power algorithm processing sub-unit communicates with the real-time processing sub-unit through
  • the isomorphic main control unit and the computing and processing sub-unit, the security processing sub-unit, the real-time processing sub-unit and the power algorithm processing sub-unit are connected by an on-chip interconnection network It interacts with the communication architecture composed of private interconnection and storage.
  • the isomorphic main control unit obtains power data and generates control instructions corresponding to power business needs by responding to business processing requests that carry power business needs.
  • the target data processing sub-unit processes the power data according to the processing request by receiving the processing request sent by the isomorphic main control unit, that is, using the same Constructive hybrid multi-chip realizes multi-core concurrent processing and multiple heterogeneous processing of power data, ensuring power data processing efficiency and power grid data security in different application scenarios.
  • a power data processing method based on a homogeneous hybrid multi-core chip is provided.
  • This embodiment uses the method applied to a terminal as an example for illustration. It can be understood that this method also It can be applied to a server, and can also be applied to a system including a terminal and a server, and is realized through the interaction between the terminal and the server.
  • the method includes the following steps:
  • Step 402 Responding to the service processing request carrying the power service demand, acquiring power data corresponding to the service processing request and generating corresponding control instructions.
  • the business processing request includes the business requirements of different power grid edge terminals, including real-time processing of power data, prediction and analysis of power grid faults, isolation or encryption of different types of power data, and power grid frequency control.
  • Step 404 sending the control instruction to the target data processing sub-module in the heterogeneous data processing module; the heterogeneous data processing module includes data processing sub-modules required by different power service requirements.
  • the heterogeneous data processing module means that each data processing sub-module in the heterogeneous data processing module adopts different types of data processors, which can meet the processing requirements of different power data in different application scenarios.
  • the heterogeneous data processing module includes power algorithm processing sub-module, real-time processing sub-module, intelligent operation processing sub-module and security processing sub-module, etc. For example, currently it is necessary to perform relay protection calculations.
  • the main control core in the main control module controls power data collection, generates and sends control instructions to the real-time calculation module at the same time, so that it can complete the corresponding protection parameter calculation. If communication is required, It is also necessary to send a control command to the security module to encrypt the protection parameter data.
  • the safety processing sub-module adopts a safety core
  • the real-time processing sub-module adopts a real-time core
  • the power algorithm processing sub-module adopts a special processor
  • the intelligent operation processing sub-module adopts an intelligent operation processor
  • the isomorphic main control module responds to the business processing request carrying the power business demand, controls the data collection equipment to collect power data, and generates control instructions;
  • the target sub-module is determined in the data processing sub-module;
  • the control instruction is used to control the data processing sub-module required by the power business demand to process the collected power data.
  • the isomorphic main control module is a high-performance module, which uses 4 identical high-performance cores.
  • Step 406 the control target data processing sub-module processes the power data according to the control instruction.
  • the isomorphic main control module in the terminal controls the security processing sub-module to perform isolation and encryption security processing on power data;
  • the module is a real-time processing sub-module
  • the real-time processing sub-module is controlled to perform real-time processing of the power data required by the power service with high real-time requirements;
  • the control The power algorithm processing sub-module determines the corresponding power algorithm to process the power data according to the processing performance requirements or data processing mode requirements; when the target data processing sub-module is determined to be an intelligent operation processing sub-module, control the intelligent operation processing sub-module Machine learning is performed on electrical data based on neural network algorithms.
  • the processing sub-module or the real-time processing sub-module controls the power algorithm processing sub-module or the real-time processing sub-module to process the power data in real time according to the control instruction.
  • the power data corresponding to the business processing request is obtained and a corresponding control instruction is generated; the control instruction is sent to the intelligence in the heterogeneous data processing module Operation processing sub-module; control the intelligent operation processing sub-module to execute the corresponding neural network algorithm according to the control instruction; perform machine learning on the power data according to the neural network algorithm to obtain implicit data representing the safety of the power grid.
  • the implicit data refers to the data of the safety status of the power grid obtained by machine learning the power data through the neural network algorithm. Extraction can be used to predict risk trends.
  • the heterogeneous data processing module includes power algorithm processing sub-module, real-time processing sub-module, intelligent operation processing sub-module and security processing sub-module, etc.
  • power algorithm processing sub-module When a business processing request is required, collect power data; when the power data is unstructured data, generate control instructions for the intelligent computing and processing sub-module, and send the control commands to the intelligent computing and processing sub-module through the Internet on chip to control the intelligent
  • the operation and processing sub-module performs machine learning on the power data according to the neural network algorithm to obtain the implicit data of the power data; Identification or power grid risk warning and other functions.
  • the sub-module processes the power data in real time, and controls the power algorithm to process the power data of the sub-module for specific processing.
  • the data required by real-time business is processed through real-time processing sub-modules, such as relay protection electrical parameter calculation (data processing, obtaining real-time electrical state quantities for monitoring and protection), real-time communication algorithm processing (obtaining sampled value messages for communications).
  • Non-real-time business data is completed through the power algorithm processing sub-module, such as the compression of data packets that need to be uploaded regularly, and the monitoring data required for power status data management.
  • the security processing sub-module encrypts the power data.
  • the target data processing sub-module is determined to process the power business requirements; the target data processing sub-module processes the power data according to the processing request by receiving the processing request sent by the isomorphic main control module, that is, the same and different Constructive hybrid multi-chip realizes multi-core concurrent processing and multiple heterogeneous processing of power data, ensuring power data processing efficiency and power grid data security in different application scenarios.
  • steps in the flow chart of FIG. 4 are displayed sequentially as indicated by the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in FIG. 4 may include multiple steps or stages, these steps or stages are not necessarily executed at the same time, but may be executed at different times, and the execution order of these steps or stages is also It is not necessarily performed sequentially, but may be performed alternately or alternately with other steps or at least a part of steps or stages in other steps.
  • Each module in the power data processing system based on the above-mentioned homogeneous hybrid multi-core chip can be realized in whole or in part by software, hardware and a combination thereof.
  • the above-mentioned modules can be embedded in or independent of the processor in the power device in the form of hardware, and can also be stored in the memory of the power device in the form of software, so that the processor can call and execute the corresponding operations of the above modules.
  • a power device is provided, the power device may be a terminal, and its internal structure may be as shown in FIG. 6 .
  • the electrical device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus.
  • the processor of the power device is used to provide calculation and control capabilities.
  • the memory of the electric device includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium stores an operating system and computer programs.
  • the internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium.
  • the communication interface of the power device is used for wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, operator network, NFC (Near Field Communication) or other technologies.
  • the computer program is executed by the processor, a power data processing method based on a heterogeneous hybrid multi-core chip is realized.
  • the display screen of the electric device may be a liquid crystal display or an electronic ink display, and the input device of the electric device may be a touch layer covered on the display screen, or a button, a trackball or a touch pad provided on the casing of the electric device .
  • FIG. 6 is only a block diagram of a part of the structure related to the solution of this application, and does not constitute a limitation on the power equipment to which the solution of this application is applied.
  • the specific power equipment can be More or fewer components than shown in the figures may be included, or some components may be combined, or have a different arrangement of components.
  • an electric device including a memory and a processor, a computer program is stored in the memory, and the processor implements the following steps when executing the computer program:
  • the heterogeneous data processing module includes data processing sub-modules required by different power business requirements
  • the control target data processing sub-module processes the power data according to the control instructions.
  • the control target data processing sub-module processes the power data according to the control instructions, including:
  • the control target data processing sub-module processes the power data according to the control instructions, including:
  • machine learning is carried out on the power data to obtain the implicit data representing the security of the power grid.
  • a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:
  • the heterogeneous data processing module includes data processing sub-modules required by different power business requirements
  • the control target data processing sub-module processes the power data according to the control instructions.
  • the control target data processing sub-module processes the power data according to the control instructions, including:
  • the control target data processing sub-module processes the power data according to the control instructions, including:
  • machine learning is carried out on the power data to obtain the implicit data representing the security of the power grid.
  • Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory or optical memory, etc.
  • Volatile memory can include Random Access Memory (RAM) or external cache memory.
  • RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Mathematical Physics (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

本申请涉及一种实现电力数据处理的同异构混合多核芯片架构。所述架构包括同构主控单元和异构数据处理单元；其中：异构数据处理单元中包括不同电力业务需求所需的数据处理子单元；同构主控单元通过响应携带电力业务需求的业务处理请求，获取电力数据以及生成电力业务需求对应的控制指令；异构数据处理单元中的目标数据处理子单元通过响应同构主控单元发送的控制指令，对电力数据进行处理。采用本系统能够提高电网边缘侧的电力数据处理效率以及安全性。

Description

一种实现电力数据处理的同异构混合多核芯片架构 技术领域
本申请涉及电网技术领域，特别是涉及一种实现电力数据处理的同异构混合多核芯片架构以及电力数据处理的方法。
背景技术
随着数字电网技术的发展，边缘计算成为承载电网业务、数据的新节点，大量数据在边缘侧实现汇聚、融合，多业务在边缘侧实现协同、就地处理。有别于传统边缘侧部署方式，数字电网边缘侧一般不形成新的物理节点，与现有装置实现融合，高并发的应用和多样化的数据处理要求需要边缘侧核心芯片具有高性能、高安全、低功耗等特征，以满足在实时响应、数据安全、智能处理的关键需求。现有边缘侧计算芯片普遍采用SMP同构多核架构，缺乏面向电网业务应用场景计算需求的异构计算单元，芯片性能及计算模式难以适应日益增长的边缘计算性能需求，缺乏边缘侧多业务数据融合的安全考虑，难以适应电网数字化的发展趋势。
目前的电力数据处理中，缺少一种融合实时、安全、智能的多元计算能力、适应于多样电力边缘计算应用场景的同/异构混合多核芯片架构，导致电网边缘侧的电力数据处理效率以及安全性较低。
发明内容
基于此，有必要针对上述技术问题，提供一种能够提高电网边缘侧的电力数据处理效率以及安全性的实现电力数据处理的同异构混合多核芯片架构以及电力数据处理方法、电力设备和存储介质可支撑实现多元电力数据在电网边缘侧的高效、安全处理。
一种实现电力数据处理的同异构混合多核芯片架构，所述同异构混合多核芯片架构包括同构主控单元和异构数据处理单元；其中：
所述异构数据处理单元中包括不同电力业务需求所需的数据处理子单元； 所述同构主控单元通过响应携带电力业务需求的业务处理请求，获取电力数据以及生成所述电力业务需求对应的控制指令；所述异构数据处理单元中的目标数据处理子单元通过接收所述同构主控单元发送的所述处理请求，根据所述处理请求对所述电力数据进行处理。
在其中一个实施例中，所述目标数据处理子单元包括智能运算处理子单元、安全处理子单元、实时处理子单元和电力算法处理子单元中至少一种；所述同构主控单元为多核同构主控单元。
在其中一个实施例中，所述安全处理子单元用于对电力数据进行隔离和加密安全处理；所述实时处理子单元用于对实时性要求高的电力业务需求的电力数据进行实时处理；所述电力算法处理子单元用于根据处理性能要求或数据处理方式要求确定对应的电力算法对所述电力数据进行处理；所述智能运算处理子单元，用于根据神经网络算法对所述电力数据进行处理。
在其中一个实施例中，所述同构主控单元采用了高性能核，所述安全处理子单元采用安全核，所述实时处理子单元采用实时核，所述电力算法处理子单元采用专用处理器，所述智能运算处理子单元采用智能运算处理器。
在其中一个实施例中，所述实时处理子单元和电力算法处理子单元之间存在私有互联，电力算法处理子单元通过私有总线与实时处理子单元进行通讯，实时处理子单元存在独立片上存储与电力算法处理子单元相连。
一种基于同异构混合多核芯片的电力数据处理方法，所述方法包括：
响应携带电力业务需求的业务处理请求，获取所述业务处理请求对应的电力数据以及生成对应的控制指令；
将所述控制指令发送至异构数据处理模组中的目标数据处理子模组；所述异构数据处理模组中包括不同电力业务需求所需的数据处理子模组；
控制所述目标数据处理子模组按照所述控制指令对所述电力数据进行处理。
在其中一个实施例中，所述目标数据处理子模组包括电力算法处理子模组和实时处理子模组；所述电力业务需求为实时性需求时，所述将所述控制指令 发送至异构数据处理模组中的目标数据处理子模组，包括：
将所述控制指令发送至异构数据处理模组中的电力算法处理子模组或实时处理子模组；
所述控制所述目标数据处理子模组按照所述控制指令对所述电力数据进行处理，包括：
控制所述电力算法处理子模组或所述实时处理子模组按照所述控制指令对所述电力数据进行实时处理。
在其中一个实施例中，所述目标数据处理子模组包括智能运算处理子模组；所述电力业务需求为非实时性需求时，所述将所述控制指令发送至异构数据处理模组中的目标数据处理子模组，包括：
将所述控制指令发送至异构数据处理模组中的智能运算处理子模组；
所述控制所述目标数据处理子模组按照所述控制指令对所述电力数据进行处理，包括：
控制所述智能运算处理子模组按照所述控制指令执行对应的神经网络算法；
根据所述神经网络算法对所述电力数据进行机器学习，得到表征电网安全的隐式数据。
所述系统架构存储器存储有计算机程序，所述处理器执行所述计算机程序时实现以下步骤：
响应携带电力业务需求的业务处理请求，获取所述业务处理请求对应的电力数据以及生成对应的控制指令；
将所述控制指令发送至异构数据处理模组中的目标数据处理子模组；所述异构数据处理模组中包括不同电力业务需求所需的数据处理子模组；
控制所述目标数据处理子模组按照所述控制指令对所述电力数据进行处理。
所述系统架构可读存储介质存储有计算机程序，所述计算机程序被处理器 执行时实现以下步骤：
响应携带电力业务需求的业务处理请求，获取所述业务处理请求对应的电力数据以及生成对应的控制指令；
将所述控制指令发送至异构数据处理模组中的目标数据处理子模组；所述异构数据处理模组中包括不同电力业务需求所需的数据处理子模组；
控制所述目标数据处理子模组按照所述控制指令对所述电力数据进行处理。
上述实现电力数据处理的同异构混合多核芯片架构，通过同构主控单元通过响应携带电力业务需求的业务处理请求，获取电力数据并生成电力业务需求对应的控制指令，从包括不同电力业务需求所需的异构数据处理单元中确定处理电力业务需求的目标数据处理子单元；目标数据处理子单元通过接收同构主控单元发送的处理请求，根据处理请求对电力数据进行处理，即采用同异构混合多芯片实现对电力数据的多核并发处理以及多元异构处理，确保不同应用场景下的电力数据处理效率以及电网数据的安全性。
附图说明
图1为一个实施例中实现电力数据处理的同异构混合多核芯片架构的示意图；
图2为另一个实施例中实现电力数据处理的同异构混合多核芯片架构的示意图；
图3为另一个实施例中实现电力数据处理的同异构混合多核芯片架构的示意图；
图4为一个实施例中基于同异构混合多核芯片架构的电力数据处理方法的流程示意图；
图5为一个实施例中基于同异构混合多核芯片架构的电力数据处理方法的应用场景图；
图6为一个实施例中电力设备的内部结构图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
在一个实施例中，如图1所示，提供了一种实现电力数据处理的同异构混合多核芯片架构，本实施例以该系统应用于终端进行举例说明，可以理解的是，该系统也可以应用于服务器。本实施例中，该系统包括同构主控单元和异构数据处理单元，其中：
同构主控单元，通过响应携带电力业务需求的业务处理请求，获取电力数据以及生成电力业务需求对应的控制指令。
其中，同构主控单元和异构数据处理单元集成在同异构混合多核芯片上；同构是指高性能主控单元中采用n(例如，4)块相同的高性能核，是同构设计，异构指多个数据处理单元采用不同类型的计算单元，可以满足不同电力数据的不同处理需求。同构主控单元和异构数据处理单元是基于通信架构进行数据交互；通信架构由片上互联网络和私有的互联和存储构成，其中，片上互联网络应用在同异构混合多核芯片的各单元之间，私有互联仅存在于特定数据处理子单元之间进行通讯(例如，实时处理子单元和电力算法子单元)。电力业务需求为电网边缘端的业务需求(例如，在SCADA系统、EMS、变电站自动化系统、换流站计算机监控系统、发电厂计算机监控系统、配电自动化系统、微机继电保护和安全自动装置中存在不同的业务需求)，业务处理请求包括对电力数据进行实时处理、对电网的故障进行预测和分析、对不同类型的电力数据进行隔离或加密以及电网频率控制等。
例如，在一个实施例中，当前需要进行继电保护运算，同构主控单元的主控核会控制进行数据采集、同时发送控制指令给异构数据处理单元中的实时处理子单元，控制实时处理子单元完成对应的保护参量计算；当需要进行通信时，对安全处理子单元发送控制指令，控制安全处理子单元对保护参量数据进行加密，将得到的加密数据进行通信。
异构数据处理单元中包括不同电力业务需求所需的数据处理子单元(如数据处理子单元1、数据处理子单元2、…..数据处理子单元n)；异构数据处理单元中的目标数据处理子单元通过接收同构主控单元发送的处理请求，根据处理请求对电力数据进行处理。
其中，异构数据处理单元包括智能运算处理子单元、安全处理子单元、实时处理子单元和电力算法处理子单元等。电力算法处理子单元是指针对电网特点或需求提出的一类应用在电力场景下的算法处理，包括前置数据处理类算法、电气参量计算类算法、数据管理类算法和网络通信类算法等；前置数据处理以及参量计算具有明确的靶向性，即可以对电力所需数据、参量、数据管理模式以及运算要求进行针对性处理；网络通信类算法有电力标准的规范约束，适合使用专用处理器来实现加速。
上述实现电力数据处理的同异构混合多核芯片架构中，同构主控单元通过响应携带电力业务需求的业务处理请求，获取电力数据以及生成电力业务需求对应的控制指令，异构数据处理单元中的目标数据处理子单元通过响应同构主控单元发送的控制指令，对电力数据进行处理，即同构主控单元用于实现多个异构数据处理单元之间的任务调度，处理控制指令和控制指令执行流程，即基于多种单元配置的同异构混合多核芯片，提高芯片异构计算能力，确保芯片能够满足在电力终端海量数据和开放互动网络环境下芯片级高性能、低时延和智能的多样性计算需求，提高电网边缘侧的电力数据处理效率以及安全性。
在一个实施例中，如图2所示，提供了一种实现电力数据处理的同异构混合多核芯片架构，本实施例以该系统应用于终端进行举例说明，可以理解的是，该方法也可以应用于服务器。本实施例中，包括同构主控单元和异构数据处理单元；其中：
异构数据处理单元中包括智能运算处理子单元、安全处理子单元、实时处理子单元和电力算法处理子单元等；同构主控单元通过响应携带电力业务需求的业务处理请求，获取电力数据以及生成电力业务需求对应的控制指令；异构数据处理单元中的目标数据处理子单元通过响应同构主控单元发送的控制指令，对电力数据进行处理。
安全处理子单元用于对电力数据进行隔离和加密安全处理；实时处理子单元用于对实时性要求高的电力业务需求的电力数据进行实时处理；电力算法处理子单元用于根据处理性能要求或数据处理方式要求确定对应的电力算法对电力数据进行处理；智能运算处理子单元用于根据神经网络算法对电力数据进行处理，得到表征电网安全的隐式数据。
不同单元之间采用了不同的处理器，同构主控单元为多核同构主控单元，同构主控单元采用了高性能核，以及多核同构架构；安全处理子单元采用安全核；实时处理子单元采用实时核；电力算法处理子单元采用专用处理器；智能运算处理子单元采用智能运算处理器。各个处理器核心针对不同业务分别单独的运算处理，彼此之间互不干扰，当一个处理器核心出现故障时，不会对其他处理器核心造成故障，提高了电力数据的处理效率以及确保电网的安全性。
同构主控单元和智能运算处理子单元、安全处理子单元、实时处理子单元和电力算法处理子单元之间通过片上互联网进行通讯，即采用可调度、可抢占的片上互联网络联通了系统中的所有单元，包括高性能核、安全核、实时核、智能运算处理器之间的通信，提高了电力数据处理效率以及通信资源的合理利用率。
片上互联网络中包括邮箱(mailbox)、中断(intc)，保证高性能核、安全核、实时核、智能运算处理器之间的信息通信，每个内核都可以触发对方的一个中断源，通过准备数据-触发中断的方式进行通信，同时，还包括片上共享内存和主存模块，用于上述核间进行数据或消息的传递；其中，Mailbox是多核芯片上用于核间通信的外设模块，可被多核调用、设置多个通道、能够接收中断信号。通过互发mail，可实现中断通知或少量数据的高速传递。
实现电力数据处理的同异构混合多核芯片架构还包括私有互联和存储。私有互联和存储存在于实时处理子单元和电力算法处理子单元构成的实时子系统中，电力算法处理子系统通过私有的总线与实时计算模块的实时核连通，确保电力业务处理请求可以被快速响应，实时计算模块中存在私有的片上存储单元与电力算法处理子系统相连，用于两个模块之间的数据不受干扰的快速传递。
利用可调度、可抢占的片上互联网络以及私有互联和存储，可以实现实时 系统之间、非实时系统之间、实时系统与非实时系统之间的数据交互以及通信，保证实时业务的及时响应和非实时业务的高效处理，保障了同/异构混合多核系统的效率以及稳定性。
可选地，在一个实施例中，主控单元采用了基于国产指令集的CPU，可以实现国产自主可控，另外，电力算法处理子单元可以实现算法种类高度定制化，根据多样的电力应用场景，例如故障诊断、状态监测、负荷辨识、停电识别、负荷控制等常用算法，设计集成不同的运算部分，满足对应需求功能，真正实现电力专用。安全处理子单元可以融合实现多种数据加密算法和安全防护算法，在边缘侧实现安全防御与攻击免疫。智能运算模块采用了低功耗、可编程神经网络处理器，适用于电力终端智能分析与处理。
在一个实施例中，如图3所示为一个实施例中实现电力数据处理的同异构混合多核芯片架构，其中，通信架构由片上互联网络和私有的互联和存储构成，同构主控单元为多核同构主控单元，安全处理子单元采用安全核，实时处理子单元采用实时核，电力算法处理子单元采用专用处理器，智能运算处理子单元采用智能运算处理器；同构主控单元和智能运算处理子单元、安全处理子单元、实时处理子单元和电力算法处理子单元之间基于片上互联网络进行通讯，即实时处理子单元和电力算法处理子单元之间存在私有互联，实时处理子单元和电力算法处理子单元之间基于私有互联进行通讯；电力算法处理子单元通过私有总线与实时处理子单元进行通讯，实时处理子单元存在独立片上存储与电力算法处理子单元相连，同构主控单元通过响应携带电力业务需求的业务处理请求，获取电力数据以及生成电力业务需求对应的控制指令；智能运算处理子单元、安全处理子单元、实时处理子单元和电力算法处理子单元中任意一个单元通过响应同构主控单元发送的控制指令，对电力数据进行处理。
上述实现电力数据处理的同异构混合多核芯片架构中，同构主控单元和能运算处理子单元、安全处理子单元、实时处理子单元和电力算法处理子单元之间，通过由片上互联网络和私有的互联和存储构成的通信架构进行数据交互，同构主控单元通过响应携带电力业务需求的业务处理请求，获取电力数据并生成电力业务需求对应的控制指令，从包括不同电力业务需求所需的异构数据处 理单元中确定处理电力业务需求的目标数据处理子单元；目标数据处理子单元通过接收同构主控单元发送的处理请求，根据处理请求对电力数据进行处理，即采用同异构混合多芯片实现对电力数据的多核并发处理以及多元异构处理，确保不同应用场景下的电力数据处理效率以及电网数据的安全性。
在一个实施例中，如图4所示，提供了一种基于同异构混合多核芯片的电力数据处理方法，本实施例以该方法应用于终端进行举例说明，可以理解的是，该方法也可以应用于服务器，还可以应用于包括终端和服务器的系统，并通过终端和服务器的交互实现。本实施例中，该方法包括以下步骤：
步骤402，响应携带电力业务需求的业务处理请求，获取业务处理请求对应的电力数据以及生成对应的控制指令。
其中，业务处理请求包括不同电网边缘端的业务需求，即包括对电力数据进行实时处理、对电网的故障进行预测和分析、对不同类型的电力数据进行隔离或加密以及电网频率控制。
步骤404，将控制指令发送至异构数据处理模组中的目标数据处理子模组；异构数据处理模组中包括不同电力业务需求所需的数据处理子模组。
其中，异构数据处理模组是指异构数据处理模组中的各数据处理子模组采用不同类型的数据处理器，可以满足不同应用场景中不同电力数据的处理需求。异构数据处理模组中包括电力算法处理子模组、实时处理子模组、智能运算处理子模组和安全处理子模组等。例如，当前需要进行继电保护运算，主控模组中的主控核控制进行电力数据采集，并生成、同时发送控制指令给实时计算模块，使其完成对应的保护参量计算，如果需要通信，则还要对安全模块发送控制指令对保护参量数据进行加密。
安全处理子模组采用安全核，实时处理子模组采用实时核，电力算法处理子模组采用专用处理器，智能运算处理子模组采用智能运算处理器。
具体地，同构主控模组响应携带电力业务需求的业务处理请求，控制数据采集设备采集电力数据，并生成控制指令；根据采集的电力数据的类型或电力业务需求从异构数据处理模组的数据处理子模组中确定目标子模组；该控制指令用于控制电力业务需求所需的数据处理子模组对采集的电力数据进行处理。 同构主控模组为高性能模组，采用4块相同的高性能核。
步骤406，控制目标数据处理子模组按照控制指令对电力数据进行处理。
具体地，当确定目标数据处理子模组为安全处理子模组时，终端中的同构主控模组控制安全处理子模组对电力数据进行隔离和加密安全处理；当确定目标数据处理子模组为实时处理子模组时，控制实时处理子模组对实时性要求高的电力业务需求的电力数据进行实时处理；当确定目标数据处理子模组为电力算法处理子模组时，控制电力算法处理子模组对处理性能要求或数据处理方式要求确定对应的电力算法对电力数据进行处理；当确定目标数据处理子模组为智能运算处理子模组时，控制智能运算处理子模组根据神经网络算法对电力数据进行机器学习。
可选地，在一个实施例中，响应携带实时性需求的业务处理请求，获取业务处理请求对应的电力数据以及生成对应的控制指令，将控制指令发送至异构数据处理模组中的电力算法处理子模组或实时处理子模组，控制电力算法处理子模组或实时处理子模组按照控制指令对电力数据进行实时处理。
可选地，在一个实施例中，响应携带非实时性需求的业务处理请求，获取业务处理请求对应的电力数据以及生成对应的控制指令；将控制指令发送至异构数据处理模组中的智能运算处理子模组；控制智能运算处理子模组按照控制指令执行对应的神经网络算法；根据神经网络算法对电力数据进行机器学习，得到表征电网安全的隐式数据。
其中，隐式数据是指通过神经网络算法对电力数据进行机器学习，得到的电网安全状况的数据，例如，根据得到的周围环境图片，判断当前电网周围是否安全，根据终端采集的电力数据进行特征提取，可以进行风险趋势预测。
以下为一个实施例中基于同异构混合多核芯片的电力数据处理方法的应用场景图，如图5所示，同异构混合多核芯片中集成了同构主控模组和异构数据处理模组，异构数据处理模组中包括电力算法处理子模组、实时处理子模组、智能运算处理子模组和安全处理子模组等，当终端中的同构主控模组响应携带电力业务需求的业务处理请求时，采集电力数据；当电力数据为非结构化数据时，生成智能运算处理子模组的控制指令，通过片上互联网将控制指令发送给 智能运算处理子模组，控制智能运算处理子模组根据神经网络算法对电力数据进行机器学习，获取电力数据的隐式数据；例如，智能运算处理子模组处理现场图像、影音等多媒体数据形式的电力数据，实现电网周围环境智能识别或者电网风险预警等功能。
当电力数据为结构化数据时，生成电力算法处理子模组、实时处理子模组的控制指令，通过片上互联网将控制指令发送给电力算法处理子模组、实时处理子模组，控制实时处理子模组对电力数据进行实时处理，控制电力算法处理子模组电力数据进行特定处理。例如，实时业务需要的数据通过实时处理子模组，比如继电保护电气参量计算(数据处理，获得实时电气状态量用于监控、保护)、实时通信类算法处理(获得采样值报文用于通信)。非实时业务数据通过电力算法处理子模组完成，比如需要定期上传的数据报文的压缩、电力状态数据管理需要的监测数据。
当业务处理请求为安全请求时，安全处理子模组对电力数据进行加密处理。
上述基于同异构混合多核芯片的电力数据处理方法中，通过响应携带电力业务需求的业务处理请求，获取电力数据并生成电力业务需求对应的控制指令，从包括不同电力业务需求所需的异构数据处理模组中确定处理电力业务需求的目标数据处理子模组；目标数据处理子模组通过接收同构主控模组发送的处理请求，根据处理请求对电力数据进行处理，即采用同异构混合多芯片实现对电力数据的多核并发处理以及多元异构处理，确保不同应用场景下的电力数据处理效率以及电网数据的安全性。
应该理解的是，虽然图4的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，图4中的至少一部分步骤可以包括多个步骤或者多个阶段，这些步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤中的步骤或者阶段的至少一部分轮流或者交替地执行。
关于基于同异构混合多核芯片的电力数据处理方法的具体限定可以参见上文中对于基于同异构混合多核芯片的电力数据处理系统的限定，在此不再赘述。上述基于同异构混合多核芯片的电力数据处理系统中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于电力设备中的处理器中，也可以以软件形式存储于电力设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。
在一个实施例中，提供了一种电力设备，该电力设备可以是终端，其内部结构图可以如图6所示。该电力设备包括通过系统总线连接的处理器、存储器、通信接口、显示屏和输入装置。其中，该电力设备的处理器用于提供计算和控制能力。该电力设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统和计算机程序。该内存储器为非易失性存储介质中的操作系统和计算机程序的运行提供环境。该电力设备的通信接口用于与外部的终端进行有线或无线方式的通信，无线方式可通过WIFI、运营商网络、NFC(近场通信)或其他技术实现。该计算机程序被处理器执行时以实现一种基于同异构混合多核芯片的电力数据处理方法。该电力设备的显示屏可以是液晶显示屏或者电子墨水显示屏，该电力设备的输入装置可以是显示屏上覆盖的触摸层，也可以是电力设备外壳上设置的按键、轨迹球或触控板。
本领域技术人员可以理解，图6中示出的结构，仅仅是与本申请方案相关的部分结构的框图，并不构成对本申请方案所应用于其上的电力设备的限定，具体的电力设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
在一个实施例中，提供了一种电力设备，包括存储器和处理器，存储器中存储有计算机程序，该处理器执行计算机程序时实现以下步骤：
响应携带电力业务需求的业务处理请求，获取业务处理请求对应的电力数据以及生成对应的控制指令；
将控制指令发送至异构数据处理模组中的目标数据处理子模组；异构数据处理模组中包括不同电力业务需求所需的数据处理子模组；
控制目标数据处理子模组按照控制指令对电力数据进行处理。
在一个实施例中，处理器执行计算机程序时还实现以下步骤：
将控制指令发送至异构数据处理模组中的电力算法处理子模组或实时处理子模组；
控制目标数据处理子模组按照控制指令对电力数据进行处理，包括：
控制电力算法处理子模组或实时处理子模组按照控制指令对电力数据进行实时处理。
在一个实施例中，处理器执行计算机程序时还实现以下步骤：
将控制指令发送至异构数据处理模组中的智能运算处理子模组；
控制目标数据处理子模组按照控制指令对电力数据进行处理，包括：
控制智能运算处理子模组按照控制指令执行对应的神经网络算法；
根据神经网络算法对电力数据进行机器学习，得到表征电网安全的隐式数据。
在一个实施例中，提供了一种计算机可读存储介质，其上存储有计算机程序，计算机程序被处理器执行时实现以下步骤：
响应携带电力业务需求的业务处理请求，获取业务处理请求对应的电力数据以及生成对应的控制指令；
将控制指令发送至异构数据处理模组中的目标数据处理子模组；异构数据处理模组中包括不同电力业务需求所需的数据处理子模组；
控制目标数据处理子模组按照控制指令对电力数据进行处理。
在一个实施例中，计算机程序被处理器执行时还实现以下步骤：
将控制指令发送至异构数据处理模组中的电力算法处理子模组或实时处理子模组；
控制目标数据处理子模组按照控制指令对电力数据进行处理，包括：
控制电力算法处理子模组或实时处理子模组按照控制指令对电力数据进行实时处理。
在一个实施例中，计算机程序被处理器执行时还实现以下步骤：
将控制指令发送至异构数据处理模组中的智能运算处理子模组；
控制目标数据处理子模组按照控制指令对电力数据进行处理，包括：
控制智能运算处理子模组按照控制指令执行对应的神经网络算法；
根据神经网络算法对电力数据进行机器学习，得到表征电网安全的隐式数据。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一非易失性计算机可读取存储介质中，该计算机程序在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用，均可包括非易失性和易失性存储器中的至少一种。非易失性存储器可包括只读存储器(Read-Only Memory，ROM)、磁带、软盘、闪存或光存储器等。易失性存储器可包括随机存取存储器(Random Access Memory，RAM)或外部高速缓冲存储器。作为说明而非局限，RAM可以是多种形式，比如静态随机存取存储器(Static Random Access Memory，SRAM)或动态随机存取存储器(Dynamic Random Access Memory，DRAM)等。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对发明专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请专利的保护范围应以所附权利要求为准。

Claims (10)

  1. 一种实现电力数据处理的同异构混合多核芯片架构，其特征在于，所述架构包括同构主控单元和异构数据处理单元；其中：
    所述异构数据处理单元中包括不同电力业务需求所需的数据处理子单元；所述同构主控单元通过响应携带电力业务需求的业务处理请求，获取电力数据以及生成所述电力业务需求对应的控制指令；所述异构数据处理单元中的目标数据处理子单元通过响应所述同构主控单元发送的所述控制指令，对所述电力数据进行处理。
  2. 根据权利要求1所述的同异构混合多核芯片架构，其特征在于，所述目标数据处理子单元包括智能运算处理子单元、安全处理子单元、实时处理子单元和电力算法处理子单元中至少一种；所述同构主控单元为多核同构主控单元。
  3. 根据权利要求2所述的同异构混合多核芯片架构，其特征在于，所述安全处理子单元用于对电力数据进行隔离和加密安全处理；所述实时处理子单元用于对实时性要求高的电力业务需求的电力数据进行实时处理；所述电力算法处理子单元用于根据处理性能要求或数据处理方式要求确定对应的电力算法对所述电力数据进行处理；所述智能运算处理子单元用于根据神经网络算法对所述电力数据进行处理。
  4. 根据权利要求2所述的同异构混合多核芯片架构，其特征在于，所述同构主控单元采用了高性能核，所述安全处理子单元采用安全核，所述实时处理子单元采用实时核，所述电力算法处理子单元采用专用处理器，所述智能运算处理子单元采用智能运算处理器。
  5. 根据权利要求3所述的同异构混合多核芯片架构，其特征在于，实时处理子单元和电力算法处理子单元之间存在私有互联，电力算法处理子单元通过私有总线与实时处理子单元进行通讯，实时处理子单元存在独立存储与电力算法处理子单元相连。
  6. 一种基于同异构混合多核芯片架构的电力数据处理方法，其特征在于，所述方法包括：
    响应携带电力业务需求的业务处理请求，获取所述业务处理请求对应的电力数据以及生成对应的控制指令；
    将所述控制指令发送至异构数据处理模组中的目标数据处理子模组；所述异构数据处理模组中包括不同电力业务需求所需的数据处理子模组；
    控制所述目标数据处理子模组按照所述控制指令对所述电力数据进行处理。
  7. 根据权利要求6所述的方法，其特征在于，所述目标数据处理子模组包括电力算法处理子模组和实时处理子模组；所述电力业务需求为实时性需求时，所述将所述控制指令发送至异构数据处理模组中的目标数据处理子模组，包括：
    将所述控制指令发送至异构数据处理模组中的电力算法处理子模组或实时处理子模组；
    所述控制所述目标数据处理子模组按照所述控制指令对所述电力数据进行处理，包括：
    控制所述电力算法处理子模组或所述实时处理子模组按照所述控制指令对所述电力数据进行实时处理。
  8. 根据权利要求6所述的方法，其特征在于，所述目标数据处理子模组包括智能运算处理子模组；所述电力业务需求为非实时性需求时，所述将所述控制指令发送至异构数据处理模组中的目标数据处理子模组，包括：
    将所述控制指令发送至异构数据处理模组中的智能运算处理子模组；
    所述控制所述目标数据处理子模组按照所述控制指令对所述电力数据进行处理，包括：
    控制所述智能运算处理子模组按照所述控制指令执行对应的神经网络算法；
    根据所述神经网络算法对所述电力数据进行机器学习，得到表征电网安全的隐式数据。
  9. 一种电力设备，包括存储器和处理器，所述存储器存储有计算机程序，其特征在于，所述处理器执行所述计算机程序时实现权利要求6至8中任一项所述的方法的步骤。
  10. 一种计算机可读存储介质，其上存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现权利要求6至8中任一项所述的方法的步骤。
PCT/CN2022/078178 2021-06-02 2022-02-28 一种实现电力数据处理的同异构混合多核芯片架构 WO2022252717A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110616472.X 2021-06-02
CN202110616472.XA CN113297126A (zh) 2021-06-02 2021-06-02 一种实现电力数据处理的同异构混合多核芯片架构

Publications (1)

Publication Number Publication Date
WO2022252717A1 true WO2022252717A1 (zh) 2022-12-08

Family

ID=77326895

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/078178 WO2022252717A1 (zh) 2021-06-02 2022-02-28 一种实现电力数据处理的同异构混合多核芯片架构

Country Status (2)

Country Link
CN (1) CN113297126A (zh)
WO (1) WO2022252717A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117318173A (zh) * 2023-09-27 2023-12-29 南方电网数字电网研究院有限公司 一种应用于大功率级联逆变器的多核异构芯片

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113297126A (zh) * 2021-06-02 2021-08-24 南方电网数字电网研究院有限公司 一种实现电力数据处理的同异构混合多核芯片架构
CN115062353B (zh) * 2022-08-16 2022-11-11 南方电网数字电网研究院有限公司 基于单芯片的可信计算系统和可信计算方法
CN115129479A (zh) * 2022-08-24 2022-09-30 南方电网数字电网研究院有限公司 电力芯片边缘数据分区处理方法、装置和计算机设备

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090070553A1 (en) * 2007-09-12 2009-03-12 Convey Computer Dispatch mechanism for dispatching insturctions from a host processor to a co-processor
CN108121683A (zh) * 2017-11-08 2018-06-05 国电南瑞科技股份有限公司 基于多核架构高速差分总线的无缝切换快速自恢复方法
CN112540951A (zh) * 2020-12-01 2021-03-23 南方电网数字电网研究院有限公司 一种适用于电力系统控制保护装置的专用主控芯片
CN112613691A (zh) * 2020-11-09 2021-04-06 贵州电网有限责任公司 芯片化继电保护通用装置
CN113297126A (zh) * 2021-06-02 2021-08-24 南方电网数字电网研究院有限公司 一种实现电力数据处理的同异构混合多核芯片架构

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7694107B2 (en) * 2005-08-18 2010-04-06 Hewlett-Packard Development Company, L.P. Dynamic performance ratio proportionate distribution of threads with evenly divided workload by homogeneous algorithm to heterogeneous computing units
US9507640B2 (en) * 2008-12-16 2016-11-29 International Business Machines Corporation Multicore processor and method of use that configures core functions based on executing instructions
CN111381885B (zh) * 2018-12-29 2024-04-09 畅想芯科有限公司 非对称多核异构并行处理系统
CN111262723B (zh) * 2020-01-07 2023-01-10 南京佶合信息科技有限公司 一种基于模块化硬件和软件定义的边缘智能计算平台

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090070553A1 (en) * 2007-09-12 2009-03-12 Convey Computer Dispatch mechanism for dispatching insturctions from a host processor to a co-processor
CN108121683A (zh) * 2017-11-08 2018-06-05 国电南瑞科技股份有限公司 基于多核架构高速差分总线的无缝切换快速自恢复方法
CN112613691A (zh) * 2020-11-09 2021-04-06 贵州电网有限责任公司 芯片化继电保护通用装置
CN112540951A (zh) * 2020-12-01 2021-03-23 南方电网数字电网研究院有限公司 一种适用于电力系统控制保护装置的专用主控芯片
CN113297126A (zh) * 2021-06-02 2021-08-24 南方电网数字电网研究院有限公司 一种实现电力数据处理的同异构混合多核芯片架构

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117318173A (zh) * 2023-09-27 2023-12-29 南方电网数字电网研究院有限公司 一种应用于大功率级联逆变器的多核异构芯片

Also Published As

Publication number Publication date
CN113297126A (zh) 2021-08-24

Similar Documents

Publication Publication Date Title
WO2022252717A1 (zh) 一种实现电力数据处理的同异构混合多核芯片架构
WO2021190482A1 (zh) 算力处理的网络系统及算力处理方法
CN112799789B (zh) 一种节点集群管理方法、装置、设备及存储介质
WO2017036244A1 (zh) 配电网数字仿真系统
US11321139B2 (en) Streaming traffic pattern for public cloud auto scaling
CN109980776A (zh) 一种智能配变系统和该系统的应用方法
WO2013104217A1 (zh) 基于云基础设施的针对应用系统维护部署的管理系统和方法
Raeisi-Varzaneh et al. Resource scheduling in edge computing: Architecture, taxonomy, open issues and future research directions
CN115063052B (zh) 一种基于电能计量芯片的电力数据处理方法和计算机设备
CN110851252A (zh) 一种基于tsn架构的规约转换设备和规约转换方法
US20230137879A1 (en) In-flight incremental processing
CN107924332A (zh) Ict服务供应的方法和系统
CN113012414A (zh) 一种支持户表停电信息实时上报的模组化能源采控终端
US11183872B2 (en) System and method for group control of distributed energy storage devices
Bumgardner et al. Cresco: A distributed agent-based edge computing framework
CN114584614A (zh) 用于智能终端与采集终端之间的信息交互方法
CN105205735A (zh) 电力调度数据云服务系统及实现方法
CN111953771A (zh) 一种配电物联网管理系统
CN112039985A (zh) 一种异构云管理方法及系统
CN109525443B (zh) 分布式前置采集通讯链路的处理方法、装置和计算机设备
CN114567568B (zh) 基于边缘计算的电力物联网数据处理方法和装置
KR101962409B1 (ko) 메시지 큐를 이용한 프로세스 간 통신을 구현하는 전력시스템
CN113905094B (zh) 工业互联网集成方法、装置及系统
CN113792934B (zh) 一种光伏发电电力并网系统、方法、设备及介质
CN112165529A (zh) 一种低成本跨网络数据交换的方法、装置、设备和介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22814765

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22814765

Country of ref document: EP

Kind code of ref document: A1