WO2023082692A1 - 一种基于业务感知的转接设备低功耗控制方法及转接设备 - Google Patents

一种基于业务感知的转接设备低功耗控制方法及转接设备 Download PDF

Info

Publication number
WO2023082692A1
WO2023082692A1 PCT/CN2022/106908 CN2022106908W WO2023082692A1 WO 2023082692 A1 WO2023082692 A1 WO 2023082692A1 CN 2022106908 W CN2022106908 W CN 2022106908W WO 2023082692 A1 WO2023082692 A1 WO 2023082692A1
Authority
WO
WIPO (PCT)
Prior art keywords
switching device
neural network
load
cyclic neural
data
Prior art date
Application number
PCT/CN2022/106908
Other languages
English (en)
French (fr)
Inventor
孙严智
陈龙
郁松
付诚
刘宇明
崔晨
彭太维
Original Assignee
云南电网有限责任公司
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 云南电网有限责任公司 filed Critical 云南电网有限责任公司
Publication of WO2023082692A1 publication Critical patent/WO2023082692A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q10/00Administration; Management
    • G06Q10/06Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063Operations research, analysis or management
    • G06Q10/0631Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06315Needs-based resource requirements planning or analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/045Combinations of networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06Energy or water supply

Definitions

  • the invention belongs to the technical field of smart grids, and in particular relates to a service-aware-based switching device low-power control method and the switching device.
  • the purpose of the present invention is to solve the deficiencies of the prior art, and provide a low-power control method for switching equipment based on service awareness. Computing and processing requirements, so as to determine its working status, and then execute the corresponding work plan to meet the requirements for low power consumption control of the switching equipment.
  • a method for controlling low power consumption of switching devices based on service awareness comprising the following steps:
  • Construct and train a bidirectional cyclic neural network input the characteristic value of the business data of the switching device into the bidirectional cyclic neural network, the bidirectional cyclic neural network outputs the business type corresponding to the business data, and calculate the The two-way cyclic neural network outputs the total value of the processing operation requirements corresponding to the business type, and compares it with the judgment criteria of the switching device in the high-load work, low-load work, and dormant state, determines its working state, and then executes the corresponding power control scheme.
  • the judging standard of the switching device working at a high load is [50%, 100%] at full load
  • the judging standard of the switching device working at a low load is [10% at full load]. %, 50%)
  • the criterion for judging that the transfer device is in a dormant state is [0%, 10%) of full load.
  • the specific method for constructing the characteristic database of various business types of the switching equipment is as follows: through parsing and processing the message data corresponding to various business types, its characteristic value can be obtained, and each business type obtains a characteristic value database.
  • the specific process of training the bidirectional recurrent neural network is as follows:
  • Intercept sample data the sample data includes training data and verification data, analyze the sample data to obtain its feature value, and manually label the feature value of the business type; when training the bidirectional cyclic neural network, use the feature value As an input, the business type corresponding to the manually marked feature value is used as an output;
  • the type of the characteristic value of the service type to be forwarded is manually marked as 1, and the type of the characteristic value of the service type to be processed is manually marked as 0; the preset number of times is 1000; the preset threshold is 0.9.
  • the present invention also provides a switching device, including a data module, a working state determination module, and a power control module;
  • the data module is used to construct the characteristic database of various business types of the switching equipment, and summarize the processing operation requirements of the various business types into a table;
  • the working state determination module is used to construct and train a bidirectional cyclic neural network, input the characteristic value of the business data of the switching device into the bidirectional cyclic neural network, and the bidirectional cyclic neural network outputs the business type corresponding to the business data, Calculate the total value of the processing operation requirements corresponding to the bidirectional cyclic neural network output business type in the same period, and compare it with the judgment criteria of the switching device in the high-load work, low-load work, and sleep states to determine its work state;
  • the power control module is used to formulate a power control scheme for the transfer device under high-load operation, low-load operation, and sleep state, and execute the corresponding power control scheme after the working state determination module determines the working state.
  • the judging standard of the switching device working at a high load is [50%, 100%] at full load
  • the judging standard of the switching device working at a low load is [10% at full load]. %, 50%)
  • the criterion for judging that the transfer device is in a dormant state is [0%, 10%) of full load.
  • the specific process of training the bidirectional recurrent neural network is as follows:
  • Intercept sample data the sample data includes training data and verification data, analyze the sample data to obtain its feature value, and manually label the feature value of the business type; when training the bidirectional cyclic neural network, use the feature value As an input, the business type corresponding to the manually marked feature value is used as an output;
  • the present invention formulates the power control scheme of the switching device under high-load operation, low-load operation, and sleep state, it can be formulated according to the existing method, and the present invention is not limited. That is, there is a power control range in different states. After setting a range here, the power control range is used to distribute the power of each component of the switching device, so as to ensure that the functions that should play a role in each state are not affected. The function of the part is enough.
  • the transfer device generally receives the data sent by the terminal. Part of these data is the data that needs to be forwarded, and the other part is the data that needs to be processed in the transfer device; here, through the messages corresponding to various business types After the data is analyzed and processed, its characteristic value can be obtained, thus forming a characteristic database.
  • the processing operation requirements of the various business types are summarized into a table, specifically: the business type to be forwarded, the memory space of the transfer device that needs to be occupied; the transfer device that needs to be processed, the transfer device that needs to be occupied The memory space of the device; and then summarize according to various business types to form a table.
  • the calculation of the total value of the processing calculation requirements corresponding to the output business types of the bidirectional cyclic neural network in the same period is specifically: the bidirectional cyclic neural network can automatically obtain the corresponding business type through the input data, and then according to the aforementioned Summarize, you can know the total value of processing operation requirements in this period.
  • the method provided by the present invention is based on the business perception of the received business data, and calculates the operation and processing requirements required by the received business type, so as to determine its working status, and then execute the corresponding working plan, so as to achieve low cost to the switching equipment. power control requirements.
  • the present invention uses a bidirectional cyclic neural network for perceptual identification of business types.
  • the bidirectional cyclic neural network has memory, parameter sharing, and Turing completeness, so it has certain advantages in learning the nonlinear characteristics of business data streams, and can Improve the accuracy of recognition.
  • FIG. 1 is a schematic flowchart of a method for controlling low power consumption of a switching device based on service perception.
  • Fig. 2 is a schematic structural diagram of the switching device.
  • plural means two or more.
  • the orientation or state relationship indicated by the terms “inner”, “upper”, “lower” and the like are based on the orientation or status relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying the No device or element must have a specific orientation, be constructed, and operate in a specific orientation and therefore should not be construed as limiting the invention.
  • connection In the description of the present invention, it should be noted that, unless otherwise specified and limited, the terms “installation”, “connection” and “installation” should be interpreted in a broad sense, for example, it can be a fixed connection or an optional connection. Detachable connection, or integral connection; it can be mechanical connection or electrical connection; it can be direct connection or indirect connection through an intermediary. Those of ordinary skill in the art will understand the specific meanings of the above terms in the present invention according to specific situations.
  • This embodiment provides a method for controlling low power consumption of a switching device based on service perception, as shown in FIG. 1 , including the following steps:
  • Construct and train a bidirectional cyclic neural network input the characteristic value of the business data of the switching device into the bidirectional cyclic neural network, the bidirectional cyclic neural network outputs the business type corresponding to the business data, and calculate the The two-way cyclic neural network outputs the total value of the processing operation requirements corresponding to the business type, and compares it with the judgment criteria of the switching device in the high-load work, low-load work, and dormant state, determines its working state, and then executes the corresponding power control scheme.
  • This embodiment provides a method for controlling low power consumption of a switching device based on service perception, as shown in FIG. 1 , including the following steps:
  • Construct and train a bidirectional cyclic neural network input the characteristic value of the business data of the switching device into the bidirectional cyclic neural network, the bidirectional cyclic neural network outputs the business type corresponding to the business data, and calculate the The two-way cyclic neural network outputs the total value of the processing operation requirements corresponding to the business type, and compares it with the judgment criteria of the switching device in the high-load work, low-load work, and dormant state, determines its working state, and then executes the corresponding power control scheme.
  • the judgment standard for the high-load operation of the switching device is [50%, 100%] of the full-load operation
  • the judgment standard of the low-load operation of the switching device is the full-load operation of [ 10%, 50%)
  • the criterion for judging that the transfer device is in a dormant state is [0%, 10%) of full load.
  • the specific method of constructing the characteristic database of various business types of the switching equipment is as follows: the characteristic value can be obtained by parsing and processing the message data corresponding to each business type, and one characteristic database is obtained for each business type.
  • the specific process of training the bidirectional recurrent neural network is as follows:
  • Intercept sample data the sample data includes training data and verification data, analyze the sample data to obtain its feature value, and manually label the feature value of the business type; when training the bidirectional cyclic neural network, use the feature value As an input, the business type corresponding to the manually marked feature value is used as an output;
  • the type of the characteristic value of the business type that needs to be forwarded is manually marked as 1, and the type of the characteristic value of the business type that needs to be processed is manually marked as 0; the preset number of times is 1000; the preset threshold is 0.9.
  • This embodiment provides a switching device, as shown in Figure 2, and its specific solution is as follows:
  • the data module is used to construct the characteristic database of various business types of the switching equipment, and summarize the processing operation requirements of the various business types into a table;
  • the working state determination module is used to construct and train a bidirectional cyclic neural network, input the characteristic value of the business data of the switching device into the bidirectional cyclic neural network, and the bidirectional cyclic neural network outputs the business type corresponding to the business data, Calculate the total value of the processing operation requirements corresponding to the bidirectional cyclic neural network output business type in the same period, and compare it with the judgment criteria of the switching device in the high-load work, low-load work, and sleep states to determine its work state;
  • the power control module is used to formulate a power control scheme for the transfer device under high-load operation, low-load operation, and sleep state, and execute the corresponding power control scheme after the working state determination module determines the working state.
  • This embodiment provides a switching device, as shown in Figure 2, and its specific solution is as follows:
  • the data module is used to construct the characteristic database of various business types of the switching equipment, and summarize the processing operation requirements of the various business types into a table;
  • the working state determination module is used to construct and train a bidirectional cyclic neural network, input the characteristic value of the business data of the switching device into the bidirectional cyclic neural network, and the bidirectional cyclic neural network outputs the business type corresponding to the business data, Calculate the total value of the processing operation requirements corresponding to the bidirectional cyclic neural network output business type in the same period, and compare it with the judgment criteria of the switching device in the high-load work, low-load work, and sleep states to determine its work state;
  • the power control module is used to formulate a power control scheme for the transfer device under high-load operation, low-load operation, and sleep state, and execute the corresponding power control scheme after the working state determination module determines the working state.
  • the judgment standard for the high-load operation of the switching device is [50%, 100%] of the full-load operation
  • the judgment standard of the low-load operation of the switching device is the full-load operation of [ 10%, 50%)
  • the criterion for judging that the transfer device is in a dormant state is [0%, 10%) of full load.
  • the specific process of training the bidirectional recurrent neural network is as follows:
  • Intercept sample data the sample data includes training data and verification data, analyze the sample data to obtain its feature value, and manually label the feature value of the business type; when training the bidirectional cyclic neural network, use the feature value As an input, the business type corresponding to the manually marked feature value is used as an output;
  • the type of the characteristic value of the business type that needs to be forwarded is manually marked as 1, and the type of the characteristic value of the business type that needs to be processed is manually marked as 0; the preset number of times is 1000; the preset threshold is 0.9.
  • the disclosed system, device and method can be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division. In actual implementation, there may be other division methods.
  • multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.
  • the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.
  • the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present invention.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical discs and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Tourism & Hospitality (AREA)
  • Software Systems (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Mathematical Physics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Quality & Reliability (AREA)
  • Development Economics (AREA)
  • Primary Health Care (AREA)
  • Educational Administration (AREA)
  • Water Supply & Treatment (AREA)
  • Public Health (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Supply And Distribution Of Alternating Current (AREA)
  • Feedback Control In General (AREA)

Abstract

本发明涉及一种基于业务感知的转接设备低功耗控制方法及转接设备,属于智能电网技术领域。该方法首先设定转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准,分别制定转接设备在三个状态下的功率控制方案;构建转接设备各种业务类型的特征数据库并将各种业务类型的处理运算要求汇总成表格;构建并训练双向循环神经网络,计算同一时期内所述双向循环神经网络输出业务类型对应的处理运算要求总值,并与所述判断标准进行比对,确定其工作状态,然后执行相应的功率控制方案。本发明基于对所接收的业务数据进行业务感知,计算接收的业务类型所需的运算处理需求,从而确定其工作状态,进而执行相应的工作方案,达到对转接设备低功耗控制的要求。

Description

一种基于业务感知的转接设备低功耗控制方法及转接设备 技术领域
本发明属于智能电网技术领域,具体涉及一种基于业务感知的转接设备低功耗控制方法及转接设备。
背景技术
随着各种通信制式的电路部件在电网领域的逐步应用,相应的转接设备如网关等也逐步在输电线路端得到了应用。然而,考虑到转接设备在输电线路侧的工作环境恶劣、取电困难,因而对其待机时间有着严格的要求,而现有的转接设备暂无相关的低功耗控制方案。因此如何克服现有技术的不足是目前智能电网技术领域亟需解决的问题。
技术问题
现有的转接设备暂无相关的低功耗控制方案。因此如何克服现有技术的不足是目前智能电网技术领域亟需解决的问题。
技术解决方案
本发明的目的是为了解决现有技术的不足,提供一种基于业务感知的转接设备低功耗控制方法,该方法基于对所接收的业务数据进行业务感知,计算接收的业务类型所需的运算处理需求,从而确定其工作状态,进而执行相应的工作方案,达到对转接设备低功耗控制的要求。
为实现上述目的,本发明采用的技术方案如下:
一种基于业务感知的转接设备低功耗控制方法,包括以下步骤:
设定转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准,分别制定转接设备在高负荷工作、低负荷工作、休眠状态下的功率控制方案;
构建转接设备各种业务类型的特征数据库,以及将所述各种业务类型的处理运算要求汇总成表格;
构建并训练一双向循环神经网络,将转接设备的业务数据的特征值输入至所述双向循环神经网络中,双向循环神经网络输出与所述业务数据对应的业务类型,计算同一时期内所述双向循环神经网络输出业务类型对应的处理运算要求总值,并与所述转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准进行比对,确定其工作状态,然后执行相应的功率控制方案。
进一步,优选的是,所述转接设备在高负荷工作的判定标准为满负荷工作的[50%,100%],所述转接设备在低负荷工作的判定标准为满负荷工作的[10%,50%),所述转接设备在休眠状态的判定标准为满负荷工作的[0%,10%)。
进一步,优选的是,构建转接设备各种业务类型的特征数据库的具体方法为:通过各种业务类型对应的报文数据进行解析处理,即可获得其特征值,每个业务类型获得一个特征数据库。
进一步,优选的是,所述训练双向循环神经网络的具体过程如下:
截取样本数据,所述样本数据包括训练数据和验证数据,对所述样本数据进行解析获取其特征值,并对所述特征值进行业务类型的人工标注;训练双向循环神经网络时,以特征值作为输入,以人工标注的特征值对应的业务类型作为输出;
将所述训练数据的特征值作为输入至所述双向循环神经网络,基于前向传播算法计算每个神经元输出值,并根据所述输出值基于反向传播算法,更新所述双向循环神经网络中的权重,重复执行上述计算每个神经元输出值并更新权重的过程直至执行总次数达到预设次数,将所述验证数据的特征输入至所述双向循环神经网络并获取识别的准确率,重复执行上述计算每个神经元输出值并更新权重直至执行总次数达到预设次数,以及获取所述准确率的过程直至获取的所述准确率大于或等于预设阈值。
进一步,优选的是,需要转发的业务类型的特征值的类型人工标记为1,需要处理的业务类型的特征值的类型人工标记为0;预设次数为1000次;预设阈值为0.9。
本发明同时提供一种转接设备,包括数据模块、工作状态确定模块、功率控制模块;
其中,数据模块用于构建转接设备各种业务类型的特征数据库,以及将所述各种业务类型的处理运算要求汇总成表格;
工作状态确定模块用于构建并训练一双向循环神经网络,将转接设备的业务数据的特征值输入至所述双向循环神经网络中,双向循环神经网络输出与所述业务数据对应的业务类型,计算同一时期内所述双向循环神经网络输出业务类型对应的处理运算要求总值,并与所述转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准进行比对,确定其工作状态;
功率控制模块用于制定转接设备在高负荷工作、低负荷工作、休眠状态下的功率控制方案,以及在工作状态确定模块确定工作状态后,执行相应的功率控制方案。
进一步,优选的是,所述转接设备在高负荷工作的判定标准为满负荷工作的[50%,100%],所述转接设备在低负荷工作的判定标准为满负荷工作的[10%,50%),所述转接设备在休眠状态的判定标准为满负荷工作的[0%,10%)。
进一步,优选的是,所述训练双向循环神经网络的具体过程如下:
截取样本数据,所述样本数据包括训练数据和验证数据,对所述样本数据进行解析获取其特征值,并对所述特征值进行业务类型的人工标注;训练双向循环神经网络时,以特征值作为输入,以人工标注的特征值对应的业务类型作为输出;
将所述训练数据的特征值作为输入至所述双向循环神经网络,基于前向传播算法计算每个神经元输出值,并根据所述输出值基于反向传播算法,更新所述双向循环神经网络中的权重,重复执行上述计算每个神经元输出值并更新权重的过程直至执行总次数达到预设次数,将所述验证数据的特征输入至所述双向循环神经网络并获取识别的准确率,重复执行上述计算每个神经元输出值并更新权重直至执行总次数达到预设次数,以及获取所述准确率的过程直至获取的所述准确率大于或等于预设阈值。
本发明制定转接设备在高负荷工作、低负荷工作、休眠状态下的功率控制方案时,按照现有方法进行制定即可,本发明不做限制。即在不同的状态下有一个功率控制范围,这里制定一个范围后,再利用所述功率控制范围对所述转接设备的各个部件的功率进行分配,确保不影响各个状态下的应发挥作用的部件的作用即可。
本发明中,转接设备一般是接收终端发过来的数据,这些数据一部分是需要转发出去的数据,另外一部分是需要在转接设备处理的数据;在这里,通过各种业务类型对应的报文数据进行解析处理,即可获得其特征值,从而形成特征数据库。
本发明中,将所述各种业务类型的处理运算要求汇总成表格,具体是:转发的业务类型,需要占用的转接设备的内存空间;需要在转接设备处理的,需要占用的转接设备的内存空间;然后按照各种业务类型进行汇总,形成表格。
本发明中,计算同一时期内所述双向循环神经网络输出业务类型对应的处理运算要求总值具体是:双向循环神经网络通过输入进来的数据,可以自动获得其对应的业务类型,然后依据前述的汇总,即可得知该时期内处理运算要求总值。
有益效果
(1)本发明提供的方法基于对所接收的业务数据进行业务感知,计算接收的业务类型所需的运算处理需求,从而确定其工作状态,进而执行相应的工作方案,达到对转接设备低功耗控制的要求。
(2)本发明应用双向循环神经网络进行业务类型的感知识别,双向循环神经网络具有记忆性、参数共享并且图灵完备,因此在对业务数据流的非线性特征进行学习时具有一定优势,可提高识别的准确性。
 
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其它的附图。
图1为基于业务感知的转接设备低功耗控制方法的流程示意图。
图2为转接设备的结构示意图。
本发明的最佳实施方式
下面结合实施例对本发明作进一步的详细描述。
本领域技术人员将会理解,下列实施例仅用于说明本发明,而不应视为限定本发明的范围。实施例中未注明具体技术或条件者,按照本领域内的文献所描述的技术或条件或者按照产品说明书进行。所用材料或设备未注明生产厂商者,均为可以通过购买获得的常规产品。
本技术领域技术人员可以理解,除非特意声明,这里使用的单数形式“一”、“一个”、“所述”和“该”也可包括复数形式。应该进一步理解的是,本发明的说明书中使用的措辞“包括”是指存在所述特征、整数、步骤、操作、元件和/或组件,但是并不排除存在或添加一个或多个其他特征、整数、步骤、操作、元件、组件和/或它们的组。应该理解,当我们称元件被“连接”到另一元件时,它可以直接连接到其他元件,或者也可以存在中间元件。此外,这里使用的“连接”可以包括无线连接。这里使用的措辞“和/或”包括一个或更多个相关联的列出项的任一单元和全部组合。
在本发明的描述中,除非另有说明,“多个”的含义是两个或两个以上。术语“内”、“上”、“下”等指示的方位或状态关系为基于附图所示的方位或状态关系,仅是为了便于描述本发明和简化描述,而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作,因此不能理解为对本发明的限制。
在本发明的描述中,需要说明的是,除非另有明确的规定和限定,术语“安装”、“连接”、“设有”应做广义理解,例如,可以是固定连接,也可以是可拆卸连接,或一体地连接;可以是机械连接,也可以是电连接;可以是直接相连,也可以通过中间媒介间接相连。对于本领域的普通技术人员而言,根据具体情况理解上述术语在本发明中的具体含义。
本技术领域技术人员可以理解,除非另外定义,这里使用的所有术语(包括技术术语和科学术语)具有与本发明所属领域中的普通技术人员的一般理解相同的意义。还应该理解的是,诸如通用字典中定义的那些术语应该被理解为具有与现有技术的上下文中的意义一致的意义,并且除非像这里一样定义,不会用理想化或过于正式的含义来解释。
实施例 1
本实施例提供了一种基于业务感知的转接设备低功耗控制方法,如图1所示,包括以下步骤:
设定转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准,分别制定转接设备在高负荷工作、低负荷工作、休眠状态下的功率控制方案;
构建转接设备各种业务类型的特征数据库,以及将所述各种业务类型的处理运算要求汇总成表格;
构建并训练一双向循环神经网络,将转接设备的业务数据的特征值输入至所述双向循环神经网络中,双向循环神经网络输出与所述业务数据对应的业务类型,计算同一时期内所述双向循环神经网络输出业务类型对应的处理运算要求总值,并与所述转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准进行比对,确定其工作状态,然后执行相应的功率控制方案。
实施例 2
本实施例提供了一种基于业务感知的转接设备低功耗控制方法,如图1所示,包括以下步骤:
设定转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准,分别制定转接设备在高负荷工作、低负荷工作、休眠状态下的功率控制方案;
构建转接设备各种业务类型的特征数据库,以及将所述各种业务类型的处理运算要求汇总成表格;
构建并训练一双向循环神经网络,将转接设备的业务数据的特征值输入至所述双向循环神经网络中,双向循环神经网络输出与所述业务数据对应的业务类型,计算同一时期内所述双向循环神经网络输出业务类型对应的处理运算要求总值,并与所述转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准进行比对,确定其工作状态,然后执行相应的功率控制方案。
在具体的实施过程中,所述转接设备在高负荷工作的判定标准为满负荷工作的[50%,100%],所述转接设备在低负荷工作的判定标准为满负荷工作的[10%,50%),所述转接设备在休眠状态的判定标准为满负荷工作的[0%,10%)。
构建转接设备各种业务类型的特征数据库的具体方法为:通过各种业务类型对应的报文数据进行解析处理,即可获得其特征值,每个业务类型获得一个特征数据库。
在具体的实施过程中,所述训练双向循环神经网络的具体过程如下:
截取样本数据,所述样本数据包括训练数据和验证数据,对所述样本数据进行解析获取其特征值,并对所述特征值进行业务类型的人工标注;训练双向循环神经网络时,以特征值作为输入,以人工标注的特征值对应的业务类型作为输出;
将所述训练数据的特征值作为输入至所述双向循环神经网络,基于前向传播算法计算每个神经元输出值,并根据所述输出值基于反向传播算法,更新所述双向循环神经网络中的权重,重复执行上述计算每个神经元输出值并更新权重的过程直至执行总次数达到预设次数,将所述验证数据的特征输入至所述双向循环神经网络并获取识别的准确率,重复执行上述计算每个神经元输出值并更新权重直至执行总次数达到预设次数,以及获取所述准确率的过程直至获取的所述准确率大于或等于预设阈值。
需要转发的业务类型的特征值的类型人工标记为1,需要处理的业务类型的特征值的类型人工标记为0;预设次数为1000次;预设阈值为0.9。
实施例 3
本实施例提供了一种转接设备,如图2所示,其具体的方案如下:
包括数据模块、工作状态确定模块、功率控制模块;
其中,数据模块用于构建转接设备各种业务类型的特征数据库,以及将所述各种业务类型的处理运算要求汇总成表格;
工作状态确定模块用于构建并训练一双向循环神经网络,将转接设备的业务数据的特征值输入至所述双向循环神经网络中,双向循环神经网络输出与所述业务数据对应的业务类型,计算同一时期内所述双向循环神经网络输出业务类型对应的处理运算要求总值,并与所述转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准进行比对,确定其工作状态;
功率控制模块用于制定转接设备在高负荷工作、低负荷工作、休眠状态下的功率控制方案,以及在工作状态确定模块确定工作状态后,执行相应的功率控制方案。
实施例 4
本实施例提供了一种转接设备,如图2所示,其具体的方案如下:
包括数据模块、工作状态确定模块、功率控制模块;
其中,数据模块用于构建转接设备各种业务类型的特征数据库,以及将所述各种业务类型的处理运算要求汇总成表格;
工作状态确定模块用于构建并训练一双向循环神经网络,将转接设备的业务数据的特征值输入至所述双向循环神经网络中,双向循环神经网络输出与所述业务数据对应的业务类型,计算同一时期内所述双向循环神经网络输出业务类型对应的处理运算要求总值,并与所述转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准进行比对,确定其工作状态;
功率控制模块用于制定转接设备在高负荷工作、低负荷工作、休眠状态下的功率控制方案,以及在工作状态确定模块确定工作状态后,执行相应的功率控制方案。
在具体的实施过程中,所述转接设备在高负荷工作的判定标准为满负荷工作的[50%,100%],所述转接设备在低负荷工作的判定标准为满负荷工作的[10%,50%),所述转接设备在休眠状态的判定标准为满负荷工作的[0%,10%)。
在具体的实施过程中,所述训练双向循环神经网络的具体过程如下:
截取样本数据,所述样本数据包括训练数据和验证数据,对所述样本数据进行解析获取其特征值,并对所述特征值进行业务类型的人工标注;训练双向循环神经网络时,以特征值作为输入,以人工标注的特征值对应的业务类型作为输出;
将所述训练数据的特征值作为输入至所述双向循环神经网络,基于前向传播算法计算每个神经元输出值,并根据所述输出值基于反向传播算法,更新所述双向循环神经网络中的权重,重复执行上述计算每个神经元输出值并更新权重的过程直至执行总次数达到预设次数,将所述验证数据的特征输入至所述双向循环神经网络并获取识别的准确率,重复执行上述计算每个神经元输出值并更新权重直至执行总次数达到预设次数,以及获取所述准确率的过程直至获取的所述准确率大于或等于预设阈值。
需要转发的业务类型的特征值的类型人工标记为1,需要处理的业务类型的特征值的类型人工标记为0;预设次数为1000次;预设阈值为0.9。
在本申请所提供的几个实施例中,应该理解到,所揭露的系统,装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本发明各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(ROM,Read-Only Memory)、随机存取存储器(RAM,Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
以上显示和描述了本发明的基本原理、主要特征和本发明的优点。本行业的技术人员应该了解,本发明不受上述实施例的限制,上述实施例和说明书中描述的只是说明本发明的原理,在不脱离本发明精神和范围的前提下,本发明还会有各种变化和改进,这些变化和改进都落入要求保护的本发明范围内。本发明要求保护范围由所附的权利要求书及其等效物界定。

Claims (8)

  1. 一种基于业务感知的转接设备低功耗控制方法,其特征在于:包括以下步骤:
    设定转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准,分别制定转接设备在高负荷工作、低负荷工作、休眠状态下的功率控制方案;
    构建转接设备各种业务类型的特征数据库,以及将所述各种业务类型的处理运算要求汇总成表格;
    构建并训练一双向循环神经网络,将转接设备的业务数据的特征值输入至所述双向循环神经网络中,双向循环神经网络输出与所述业务数据对应的业务类型,计算同一时期内所述双向循环神经网络输出业务类型对应的处理运算要求总值,并与所述转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准进行比对,确定其工作状态,然后执行相应的功率控制方案。
  2. 根据权利要求1所述的基于业务感知的转接设备低功耗控制方法,其特征在于:所述转接设备在高负荷工作的判定标准为满负荷工作的[50%,100%],所述转接设备在低负荷工作的判定标准为满负荷工作的[10%,50%),所述转接设备在休眠状态的判定标准为满负荷工作的[0%,10%)。
  3. 根据权利要求1所述的基于业务感知的转接设备低功耗控制方法,其特征在于:构建转接设备各种业务类型的特征数据库的具体方法为:通过各种业务类型对应的报文数据进行解析处理,即可获得其特征值,每个业务类型获得一个特征数据库。
  4. 根据权利要求1所述的基于业务感知的转接设备低功耗控制方法,其特征在于:所述训练双向循环神经网络的具体过程如下:
    截取样本数据,所述样本数据包括训练数据和验证数据,对所述样本数据进行解析获取其特征值,并对所述特征值进行业务类型的人工标注;训练双向循环神经网络时,以特征值作为输入,以人工标注的特征值对应的业务类型作为输出;
    将所述训练数据的特征值作为输入至所述双向循环神经网络,基于前向传播算法计算每个神经元输出值,并根据所述输出值基于反向传播算法,更新所述双向循环神经网络中的权重,重复执行上述计算每个神经元输出值并更新权重的过程直至执行总次数达到预设次数,将所述验证数据的特征输入至所述双向循环神经网络并获取识别的准确率,重复执行上述计算每个神经元输出值并更新权重直至执行总次数达到预设次数,以及获取所述准确率的过程直至获取的所述准确率大于或等于预设阈值。
  5. 根据权利要求4所述的基于业务感知的转接设备低功耗控制方法,其特征在于:需要转发的业务类型的特征值的类型人工标记为1,需要处理的业务类型的特征值的类型人工标记为0;预设次数为1000次;预设阈值为0.9。
  6. 一种转接设备,其特征在于:包括数据模块、工作状态确定模块、功率控制模块;
    其中,数据模块用于构建转接设备各种业务类型的特征数据库,以及将所述各种业务类型的处理运算要求汇总成表格;
    工作状态确定模块用于构建并训练一双向循环神经网络,将转接设备的业务数据的特征值输入至所述双向循环神经网络中,双向循环神经网络输出与所述业务数据对应的业务类型,计算同一时期内所述双向循环神经网络输出业务类型对应的处理运算要求总值,并与所述转接设备在高负荷工作、低负荷工作、休眠状态下的判断标准进行比对,确定其工作状态;
    功率控制模块用于制定转接设备在高负荷工作、低负荷工作、休眠状态下的功率控制方案,以及在工作状态确定模块确定工作状态后,执行相应的功率控制方案。
  7. 根据权利要求6所述的转接设备,其特征在于:所述转接设备在高负荷工作的判定标准为满负荷工作的[50%,100%],所述转接设备在低负荷工作的判定标准为满负荷工作的[10%,50%),所述转接设备在休眠状态的判定标准为满负荷工作的[0%,10%)。
  8. 根据权利要求6所述的转接设备,其特征在于:所述训练双向循环神经网络的具体过程如下:
    截取样本数据,所述样本数据包括训练数据和验证数据,对所述样本数据进行解析获取其特征值,并对所述特征值进行业务类型的人工标注;训练双向循环神经网络时,以特征值作为输入,以人工标注的特征值对应的业务类型作为输出;
    将所述训练数据的特征值作为输入至所述双向循环神经网络,基于前向传播算法计算每个神经元输出值,并根据所述输出值基于反向传播算法,更新所述双向循环神经网络中的权重,重复执行上述计算每个神经元输出值并更新权重的过程直至执行总次数达到预设次数,将所述验证数据的特征输入至所述双向循环神经网络并获取识别的准确率,重复执行上述计算每个神经元输出值并更新权重直至执行总次数达到预设次数,以及获取所述准确率的过程直至获取的所述准确率大于或等于预设阈值。
PCT/CN2022/106908 2021-11-09 2022-07-21 一种基于业务感知的转接设备低功耗控制方法及转接设备 WO2023082692A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111322511.1 2021-11-09
CN202111322511.1A CN114240035B (zh) 2021-11-09 2021-11-09 一种基于业务感知的转接设备低功耗控制方法及转接装置

Publications (1)

Publication Number Publication Date
WO2023082692A1 true WO2023082692A1 (zh) 2023-05-19

Family

ID=80748952

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/106908 WO2023082692A1 (zh) 2021-11-09 2022-07-21 一种基于业务感知的转接设备低功耗控制方法及转接设备

Country Status (2)

Country Link
CN (1) CN114240035B (zh)
WO (1) WO2023082692A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114240035B (zh) * 2021-11-09 2024-03-22 云南电网有限责任公司 一种基于业务感知的转接设备低功耗控制方法及转接装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103699440A (zh) * 2012-09-27 2014-04-02 北京搜狐新媒体信息技术有限公司 一种云计算平台系统为任务分配资源的方法和装置
US20190265768A1 (en) * 2018-02-24 2019-08-29 Hefei University Of Technology Method, system and storage medium for predicting power load probability density based on deep learning
CN110806918A (zh) * 2019-09-24 2020-02-18 梁伟 基于深度学习神经网络的虚拟机运行方法和装置
CN111245667A (zh) * 2018-11-28 2020-06-05 中国移动通信集团浙江有限公司 网络业务识别方法及装置
CN111629380A (zh) * 2020-05-09 2020-09-04 中国科学院沈阳自动化研究所 面向高并发多业务工业5g网络的动态资源分配方法
CN114240035A (zh) * 2021-11-09 2022-03-25 云南电网有限责任公司 一种基于业务感知的转接设备低功耗控制方法及转接设备

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110285532B (zh) * 2019-07-04 2021-07-30 中国工商银行股份有限公司 基于人工智能的机房空调控制方法、装置及系统
CN112241331A (zh) * 2020-11-19 2021-01-19 西北大学 一种基于深度神经网络的移动Web任务动态迁移方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103699440A (zh) * 2012-09-27 2014-04-02 北京搜狐新媒体信息技术有限公司 一种云计算平台系统为任务分配资源的方法和装置
US20190265768A1 (en) * 2018-02-24 2019-08-29 Hefei University Of Technology Method, system and storage medium for predicting power load probability density based on deep learning
CN111245667A (zh) * 2018-11-28 2020-06-05 中国移动通信集团浙江有限公司 网络业务识别方法及装置
CN110806918A (zh) * 2019-09-24 2020-02-18 梁伟 基于深度学习神经网络的虚拟机运行方法和装置
CN111629380A (zh) * 2020-05-09 2020-09-04 中国科学院沈阳自动化研究所 面向高并发多业务工业5g网络的动态资源分配方法
CN114240035A (zh) * 2021-11-09 2022-03-25 云南电网有限责任公司 一种基于业务感知的转接设备低功耗控制方法及转接设备

Also Published As

Publication number Publication date
CN114240035B (zh) 2024-03-22
CN114240035A (zh) 2022-03-25

Similar Documents

Publication Publication Date Title
CN1920749B (zh) 节电电子装置及其节电方法
JP2015043225A (ja) ネットワークアクセスモジュールの電源管理方法及び装置
US20150312868A1 (en) Node device and method for communication
WO2023082692A1 (zh) 一种基于业务感知的转接设备低功耗控制方法及转接设备
CN108777862B (zh) 一种蓝牙传输方法、蓝牙控制器以及蓝牙设备
CN204270291U (zh) 一种基于smc的监控报警装置
CN105676996A (zh) 一种龙芯服务器的功耗控制方法和装置
CN109388604A (zh) 一种基于PCIe的热插拔控制方法、装置及存储介质
US11804236B2 (en) Method for debugging noise elimination algorithm, apparatus and electronic device
WO2022002068A1 (zh) 数据处理方法、系统、设备及存储介质
CN111913669B (zh) Ssd掉电速度提高方法、装置、计算机设备及存储介质
WO2021103309A1 (zh) 一种Expander的带外更新方法和系统
CN104424142A (zh) 一种多核处理器系统中访问共享资源的方法与装置
CN207799397U (zh) 一种具有休眠苏醒功能的gprs型rtu
CN104834562A (zh) 一种异构数据中心及该数据中心的运行方法
CN204859227U (zh) 基于物联网的建筑结构健康监测装置
CN202854647U (zh) 零终端机
CN204595694U (zh) 一种基于申威410处理器的cpci计算机主板
CN102866732A (zh) 零终端机
CN110766566A (zh) 一种基于双向lstm模型的运维行为智能分析系统
WO2020001065A1 (zh) 一种bios中更改系统能效模式的方法、系统及服务器
LU505330B1 (en) Low power consumption control method for switching equipment based on business sensing and same
CN205754421U (zh) 用于数据集群的应用服务器终端机
CN104182030A (zh) 一种基于业务使用量的服务器节能策略方法及系统
WO2024022100A1 (zh) 数据传输方法、装置、控制设备、暖通设备及存储介质

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: 22891501

Country of ref document: EP

Kind code of ref document: A1