CN109218414A - A kind of distributed computing method of smart grid-oriented hybrid network framework - Google Patents
A kind of distributed computing method of smart grid-oriented hybrid network framework Download PDFInfo
- Publication number
- CN109218414A CN109218414A CN201810983999.4A CN201810983999A CN109218414A CN 109218414 A CN109218414 A CN 109218414A CN 201810983999 A CN201810983999 A CN 201810983999A CN 109218414 A CN109218414 A CN 109218414A
- Authority
- CN
- China
- Prior art keywords
- particle
- node
- task
- antibody
- immune
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of distributed computing methods of smart grid-oriented hybrid network framework.It is the following steps are included: establish system model；Initialize the basic parameter of immune Chaos particle swarm optimization algorithm；It is random to generate NpA particle calculates the processing delay of entire task under current task distribution condition；Calculate current NpThe fitness value of each particle in a particle；Judge particle the number of iterations, if particle the number of iterations reaches the immune interference step-length e of setting, carries out immune mutation operation, otherwise update the individual adaptive optimal control angle value of each particle and global optimum's fitness value of all particles；Judge whether the maximum number of iterations for reaching immune Chaos particle swarm optimization algorithm, if so, the optimal case that the corresponding particle of global optimum's fitness value for exporting the algorithm is distributed as task, otherwise continues iteration.The present invention can calculate the optimal case for distribution of going out on missions, and improve smart grid to the processing capacity of high concurrent, low time delay mission requirements.
Description
Technical field
The present invention relates to intelligent power grid technology field more particularly to a kind of distributions of smart grid-oriented hybrid network framework
Formula calculation method.
Background technique
The data volume generated with the raising of the grid-connected rate of distributed energy and the fast development of smart grid, electric system
Exponentially increase again, how to utilize advanced information processing system, improves the big data processing capacity and control energy of electric system
Power becomes smart grid and develops the major issue that must be faced and solve.Since renewable energy is intermittent, uncontrollable
Feature, cutting-in control bring very big difficulty to electric system, and to the requirement with higher of information processing time delay.Currently,
The concentration information processing manner of cloud computing, due to server apart from terminal device farther out, the remote transmission belt of mass data come
The problems such as biggish delay problem, bandwidth problem and control deficiency.Therefore, smart grid information processing system, which becomes, improves electric power
The major issue of quality of system control.
Summary of the invention
For the present invention in order to solve existing intelligent electrical network mass data undertreatment, smart grid data volume is huge, and network passes
Defeated bandwidth, computing capability are limited and distributed power generation brings grid power to fluctuate the technical problem high to requirement of real-time, mention
The distributed computing method for having supplied a kind of smart grid-oriented hybrid network framework can calculate the optimal side for distribution of going out on missions
Case improves smart grid to the processing capacity of high concurrent, low time delay mission requirements, task processing delay is effectively reduced, improves and supplies
Electricity quality and control ability.
To solve the above-mentioned problems, the present invention is achieved by the following scheme:
A kind of distributed computing method of smart grid-oriented hybrid network framework of the invention, the hybrid network framework
Including cloud computing layer, load balancing layer, edge calculations layer, terminal device layer and transport network layer, cloud computing layer includes a cloud
Calculate node, edge calculations layer include L edge calculations node, comprising the following steps:
Establish system model f；
The task amount that each calculate node should distribute is calculated using immune Chaos particle swarm optimization algorithm, calculate node includes cloud
Calculate node and edge calculations node, comprising the following steps:
S1: being configured the basic parameter of immune Chaos particle swarm optimization algorithm, including population scale Np, population dimension
D, maximum number of iterations T, Studying factors a1, Studying factors α2, originate weights omegas, terminate weights omegae, chaos controlling coefficient μ, mix
Ignorant variable zit, immune-regulating factor d0, immune interference step-length e；
S2: indicating the task amount of each calculate node using the position of particle, to the distributing to each calculate node of the task
Amount range is configured, and generates N at randompA particle calculates the processing delay of entire task under current task distribution condition；
S3: make the inertia weight of particle swarm algorithm in starting weights omega using the randomness of chaotic motionsWith termination weight
ωeBetween random jump, so that particle is carried out global search around optimal solution, continue open up globally optimal solution, avoid falling into office
Portion is optimal, and the chaos sequence formula of generation is as follows:
zit+1=μ zit(1-zit),
Wherein, it is current iteration number, zitThe Chaos Variable generated for i-th t times iteration；
S4: the inertia weight ω (it) mapped according to chaotic motion, formula is as follows:
S5: current N is calculatedpThe fitness value of each particle in a particle, fitness value are to use current particle corresponding
Task distribution condition under entire task processing delay, and calculate separately out the individual adaptive optimal control angle value of each particle and complete
Global optimum's fitness value of portion's particle；
S6: inertia weight ω (it) is brought into the particle rapidity more new formula of current iteration number and particle position updates
In formula, the speed and position of population are updated,
Particle rapidity more new formula is as follows:
Wherein,For i-th of particle i-th t times iteration moving step length,The history lived through for i-th of particle
Adaptive optimal control angle value, pbFor global optimum's fitness value of all particles, γ1、γ2The equally distributed random number between 0 and 1,It is i-th of particle in the position of i-th t times iteration；
Particle position more new formula is as follows:
S7: judging particle the number of iterations, if particle the number of iterations reaches the immune interference step-length e of setting, is immunized
Mutation operation continues to execute step S8, otherwise, jumps to step S12；
S8: when executing immune mutation operation, by population NpIn each particle regard an antibody, antibody, that is, task as
Sendout, then random generation NrA antibody, by population NpWith population NrMerge into a new Np+NrPopulation；
S9: the select probability of each antibody is calculated:
Wherein, P1iFor the select probability of i-th of antibody,For the fitness value of i-th of antibody,
For the affine force value of i-th of antibody；
S10: the select probability of the concentration of each antibody is calculated:
Wherein, P2iFor the select probability of the concentration of i-th of antibody,For the concentration value of i-th of antibody；
S11: the probability based on affinity and concentration selection of each antibody is calculated:
Pi=θ P1i+(1-θ)P2i,
Wherein, PiFor the probability based on affinity and concentration selection of i-th of antibody, θ is the coordination system greater than 0 less than 1
Number, for coordinating the specific gravity of affinity and concentration；
The probability value that antibody is selected according to affinity with concentration is sorted from high to low, select probability value it is highest before
NpA antibody is as new population particle, and for the optimizing of particle next time, then go to step S3；
S12: the individual adaptive optimal control angle value of each particle and global optimum's fitness value of all particles are updated；
S13: judge whether the maximum number of iterations for reaching immune Chaos particle swarm optimization algorithm, if so, exporting the algorithm
The optimal case that the corresponding particle of global optimum's fitness value is distributed as task, otherwise go to step S3.
The population diversity of population is kept using self immunity function of immune algorithm itself, meanwhile, it utilizes
The randomness of chaotic motion and the inertia weight system of particle swarm algorithm is improved, improves the convergence rate of particle swarm algorithm
And global optimizing ability.
Preferably, the method for establishing system model f the following steps are included:
Assuming that by C1、C2、C3......CLRespectively represent L edge calculations node, CcdA cloud computing node is represented,
The communication overhead between any two node is represented,
Wherein,Indicate any two nodes CIAnd CJBetween relationship,It can only be 0 or 1, when
When being 0, two node C are indicatedIAnd CJBetween there are the task relations of distribution；WhenWhen being 1, two node C are indicatedIAnd CJBetween
There is no the task relations of distribution；Then indicate any two nodes CIAnd CJBetween data transmission delay；
Assuming that the computing capability of each edge calculations node is used respectively It indicates, cloud meter
The computing capability of operator node is usedIt indicates, waiting task is indicated with U, distributes to edge calculations node and cloud computing node
Subtask ratio δIAnd δcdIt indicates, data transmission delayBy transmission delayAnd propagation delayIt indicating, wherein K is data frame length,The bandwidth of communication network between two nodes；
Since hybrid network framework uses distributed computing method, a task is distributed into multiple adjacent nodes and is located parallel
Reason, so, the processing time of that subtask of longest when the processing time of entire task is equal to the collaboration processing of all subtasks, because
This, the model foundation of whole system is as follows:
I=1,2 ..., L, J=1,2 ..., L,
Preferably, cloud computing layer provides reliable data processing and secure storage service for system；Load balancing layer exists
Task is distributed between edge calculations node and cloud computing node；Edge calculations layer provides calculating and storage service nearby for equipment；
Terminal device layer provides various supply and distribution services for system；Transport network layer provides efficient, reliable clothes for share of system information
Business.
Preferably, the transport network layer is made of low-power consumption wide area network LoRa network.
The beneficial effects of the present invention are: the optimal case for distribution of going out on missions can be calculated, improve smart grid to high concurrent,
Task processing delay is effectively reduced in the processing capacity of low time delay mission requirements, improves power supply quality and control ability.
Detailed description of the invention
Fig. 1 is flow chart of the invention；
Fig. 2 is the schematic diagram of hybrid network framework；
Fig. 3 is the undirected weight map of system；
Fig. 4 is the delay performance comparison of inventive network framework and system for cloud computing, single edge calculations node；
Fig. 5 is the method for the present invention and the delay comparison of other load-balancing methods.
Specific embodiment
Below with reference to the embodiments and with reference to the accompanying drawing the technical solutions of the present invention will be further described.
Embodiment: the distributed computing method of more equipment coordination cooperations in a kind of smart grid-oriented of the present embodiment, such as
Shown in Fig. 2, hybrid network framework includes cloud computing layer, load balancing layer, edge calculations layer, terminal device layer and network transmission
Layer, cloud computing layer include a cloud computing node, edge calculations layer includes L edge calculations node, as shown in Figure 1, include with
Lower step:
Establish system model f；
The task amount that each calculate node should distribute is calculated using immune Chaos particle swarm optimization algorithm, calculate node includes cloud
Calculate node and edge calculations node, comprising the following steps:
S1: being configured the basic parameter of immune Chaos particle swarm optimization algorithm, including population scale Np, population dimension
D, maximum number of iterations T, Studying factors a1, Studying factors a2, originate weights omegas, terminate weights omegae, chaos controlling coefficient μ, mix
Ignorant variable zit, immune-regulating factor d0, immune interference step-length e；
S2: indicating the task amount of each calculate node using the position of particle, to the distributing to each calculate node of the task
Amount range is configured, and generates N at randompA particle calculates the processing delay of entire task under current task distribution condition；
S3: make the inertia weight of particle swarm algorithm in starting weights omega using the randomness of chaotic motionsWith termination weight
ωeBetween random jump, so that particle is carried out global search around optimal solution, continue open up globally optimal solution, avoid falling into office
Portion is optimal, and the chaos sequence formula of generation is as follows:
zit+1=μ zit(1-zit),
Wherein, it is current iteration number, zitThe Chaos Variable generated for i-th t times iteration；
S4: the inertia weight ω (it) mapped according to chaotic motion, formula is as follows:
S5: current N is calculatedpThe fitness value of each particle in a particle, fitness value are to use current particle corresponding
Task distribution condition under entire task processing delay, and calculate separately out the individual adaptive optimal control angle value of each particle and complete
Global optimum's fitness value of portion's particle；
S6: inertia weight ω (it) is brought into the particle rapidity more new formula of current iteration number and particle position updates
In formula, the speed and position of population are updated,
Particle rapidity more new formula is as follows:
Wherein,For i-th of particle i-th t times iteration moving step length,The history lived through for i-th of particle
Adaptive optimal control angle value, pbFor global optimum's fitness value of all particles, γ1、γ2The equally distributed random number between 0 and 1,It is i-th of particle in the position of i-th t times iteration；
Particle position more new formula is as follows:
S7: judging particle the number of iterations, if particle the number of iterations reaches the immune interference step-length e of setting, is immunized
Mutation operation continues to execute step S8, otherwise, jumps to step S12；
S8: when executing immune mutation operation, by population NpIn each particle regard an antibody, antibody, that is, task as
Sendout, then random generation NrA antibody, by population NpWith population NrMerge into a new Np+NrPopulation；
S9: the select probability of each antibody is calculated:
Wherein, P1iFor the select probability of i-th of antibody,For the fitness value of i-th of antibody,
For the affine force value of i-th of antibody；
S10: the select probability of the concentration of each antibody is calculated:
Wherein, P2iFor the select probability of the concentration of i-th of antibody,For the concentration value of i-th of antibody；
S11: the probability based on affinity and concentration selection of each antibody is calculated:
Pi=θ P1i+(1-θ)P2i,
Wherein, PiFor the probability based on affinity and concentration selection of i-th of antibody, θ is the coordination system greater than 0 less than 1
Number, for coordinating the specific gravity of affinity and concentration；
The probability value that antibody is selected according to affinity with concentration is sorted from high to low, select probability value it is highest before
NpA antibody is as new population particle, and for the optimizing of particle next time, then go to step S3；
S12: the individual adaptive optimal control angle value of each particle and global optimum's fitness value of all particles are updated；
S13: judge whether the maximum number of iterations for reaching immune Chaos particle swarm optimization algorithm, if so, exporting the algorithm
The optimal case that the corresponding particle of global optimum's fitness value is distributed as task, otherwise go to step S3.
The population diversity of population is kept using self immunity function of immune algorithm itself, meanwhile, it utilizes
The randomness of chaotic motion and the inertia weight system of particle swarm algorithm is improved, improves the convergence rate of particle swarm algorithm
And global optimizing ability.
Establish the method for system model f the following steps are included:
Assuming that by C1、C2、C3......CLRespectively represent L edge calculations node, CcdA cloud computing node is represented,
The communication overhead between any two node is represented, as shown in figure 3,
Wherein,Indicate any two nodes CIAnd CJBetween relationship,It can only be 0 or 1, when
When being 0, two node C are indicatedIAnd CJBetween there are the task relations of distribution；WhenWhen being 1, two node C are indicatedIAnd CJBetween
There is no the task relations of distribution；Then indicate any two nodes CIAnd CJBetween data transmission delay；
Assuming that the computing capability of each edge calculations node is used respectively It indicates, cloud meter
The computing capability of operator node is usedIt indicates, waiting task is indicated with U, distributes to edge calculations node and cloud computing node
Subtask ratio δIAnd δcdIt indicates, data transmission delayBy transmission delayAnd propagation delayIt indicating, wherein K is data frame length,The bandwidth of communication network between two nodes；
Since hybrid network framework uses distributed computing method, a task is distributed into multiple adjacent nodes and is located parallel
Reason, so, the processing time of that subtask of longest when the processing time of entire task is equal to the collaboration processing of all subtasks, because
This, the model foundation of whole system is as follows:
I=1,2 ..., L, J=1,2 ..., L,
Cloud computing layer provides reliable data processing and secure storage service for system.The characteristics of its flexible, easy extension, make
User (enterprise, resident etc.) can pass through the terminal devices such as mobile phone, computer in the case where network connection, inquiry whenever and wherever possible,
Transacting business (electricity consumption inquiry, payment, Saving energy etc..
Load balancing layer distributes task between edge calculations node and cloud computing node.Load balancing layer includes main load
Balancer and backup load balancer.
Edge calculations layer provides calculating and storage service nearby for equipment.Mainly by communication base station, private server and adopt
Collect the network edge device composition with certain computing capability and storage capacity such as equipment.These edge devices close to data source,
It is directly connected to, collected information can be handled nearby by various wired, wireless modes and terminal device, and processing is tied
Fruit or important information are transferred to cloud and carry out depth excavation and storage preservation.
Terminal device layer provides various supply and distribution services for system.It is made of various controller switching equipments, generating equipment etc., such as
The equipment such as transformer, photovoltaic panel, charging pile and energy storage.Wirelessly it is connected with neighbouring edge calculations node, forms
One physical connection network extensively covered is realized to the safe and reliable service of power supply system.
Transport network layer provides efficient, reliability services for share of system information.It realizes the data transmission between each layer, makes each
It can be very good cooperation exchange between each layer of terminal device, make optimum control convenient for system.
Transport network layer is made of low-power consumption wide area network LoRa network.
The delay performance of this programme network architecture and system for cloud computing, single edge calculations node compares, as shown in Figure 4.
This programme method and the delay comparison of other load-balancing methods, as shown in Figure 5.
Since the computing capability and storage capacity of edge calculations node are limited, in order to improve task treatment effeciency, reduce system
Unite time delay, this method according to the computing capability and net processing overhead of different edge calculations nodes, task is rationally decomposed,
Distribution, parallel processing achieve the purpose that minimum task processing delay.This method can calculate the optimal side for distribution of going out on missions
Case effectively jumps out local optimum, reduces task processing delay, improves smart grid to high concurrent, low time delay mission requirements
Task processing delay is effectively reduced in processing capacity, improves power supply quality and control ability.
Since the Cloud Server of cloud computing layer is far apart from terminal device, if all data that terminal generates all uploaded
Cloud Server, it will generate biggish delay, in order to meet the requirement of system real time, data are carried out using fringe node real
When handle.Since the storage of fringe node, computing resource are limited, when there is task to reach, in order to efficiently obtain processing knot in time
Fruit needs to decompose task using load-balancing algorithm, and processing is cooperateed with other nodes, improves computing capability.Long-range prison
Some less important and simple task data is passed directly to end by load balancer by the collected information of measurement equipment
The base station LoRa, generic server near end equipment and the fringe node with calculating, storage capacity, are pre-processed, and
By some important, complex task data for needing to comprehensively consider power grid other information, be transferred to Cloud Server carry out unified storage,
Processing.
When terminal device breaks down, when needing to interact with neighbouring device, directly by the interaction of fringe node into
Row solves, and the equipment response time thus can be improved, and reduces network bandwidth and mitigates cloud computation burden.
Cloud Server is made of the high performance server of a group, using virtual technology provide powerful data-handling capacity and
Storage capacity, power consumer can obtain power information and related Saving energy whenever and wherever possible by terminal devices such as mobile phones；Match
The clients such as electric administrative staff, maintenance personal, sale of electricity company, Utilities Electric Co. can also access network whenever and wherever possible, and be taken according to cloud
Equipment optimization, the upgrading processing suggestion of business the device Visualization Service provided and push, in time, efficiently grasp electric network information, for visitor
Family offer is efficient, facilitates good grid service.Cloud Server provides unified an interface and communication protocol, facilitates different
User's access obtains different services, and user is not necessarily to pay close attention to the storage details and network service of bottom, need to only be absorbed in the industry of oneself
Business development.
Claims (5)
1. a kind of distributed computing method of smart grid-oriented hybrid network framework, the hybrid network framework includes cloud computing
Layer, load balancing layer, edge calculations layer, terminal device layer and transport network layer, cloud computing layer include a cloud computing node, side
Edge computation layer includes L edge calculations node, which comprises the following steps:
Establish system model f；
The task amount that each calculate node should distribute is calculated using immune Chaos particle swarm optimization algorithm,
Calculate node includes cloud computing node and edge calculations node.
2. a kind of distributed computing method of smart grid-oriented hybrid network framework according to claim 1, feature
It is, the method for calculating the task amount that each calculate node should distribute using immune Chaos particle swarm optimization algorithm includes following
Step:
S1: being configured the basic parameter of immune Chaos particle swarm optimization algorithm, including population scale Np, population dimension D, most
Big the number of iterations T, Studying factors a1, Studying factors a2, originate weights omegas, terminate weights omegae, chaos controlling coefficient μ, chaos change
Measure Zit, immune-regulating factor d0, immune interference step-length e；
S2: indicating the task amount of each calculate node using the position of particle, to the task amount model for distributing to each calculate node
It encloses and is configured, generate N at randompA particle calculates the processing delay of entire task under current task distribution condition；
S3: make the inertia weight of particle swarm algorithm in starting weights omega using the randomness of chaotic motionsWith termination weights omegaeIt
Between random jump, so that particle is carried out global search around optimal solution, continue to open up globally optimal solution, avoid falling into part most
Excellent, the chaos sequence formula of generation is as follows:
zit+1=μ zit(1-zit),
Wherein, it is current iteration number, zitThe Chaos Variable generated for i-th t times iteration；
S4: the inertia weight ω (it) mapped according to chaotic motion, formula is as follows:
S5: current N is calculatedpThe fitness value of each particle in a particle, fitness value are the corresponding task of current particle that uses
Entire task processing delay under distribution condition, and calculate separately out the individual adaptive optimal control angle value and all particles of each particle
Global optimum's fitness value；
S6: inertia weight ω (it) is brought into the particle rapidity more new formula and particle position more new formula of current iteration number
In, the speed and position of population are updated,
Particle rapidity more new formula is as follows:
Wherein,For i-th of particle i-th t times iteration moving step length,The history lived through for i-th of particle is optimal
Fitness value, pbFor global optimum's fitness value of all particles, γ1、γ2The equally distributed random number between 0 and 1,
It is i-th of particle in the position of i-th t times iteration；
Particle position more new formula is as follows:
S7: judging particle the number of iterations, if particle the number of iterations reaches the immune interference step-length e of setting, carries out immune variation
Operation continues to execute step S8, otherwise, jumps to step S12；
S8: when executing immune mutation operation, by population NpIn each particle regard an antibody, the distribution of antibody, that is, task as
Amount, then random generation NrA antibody, by population NpWith population NrMerge into a new Np+NrPopulation；
S9: the select probability of each antibody is calculated:
Wherein, P1iFor the select probability of i-th of antibody,For the fitness value of i-th of antibody,It is i-th
The affine force value of a antibody；
S10: the select probability of the concentration of each antibody is calculated:
Wherein, P2iFor the select probability of the concentration of i-th of antibody,For the concentration value of i-th of antibody；
S11: the probability based on affinity and concentration selection of each antibody is calculated:
Pi=θ P1i+(1-θ)P2i,
Wherein, PiFor the probability based on affinity and concentration selection of i-th of antibody, θ is the cooperation index greater than 0 less than 1, is used
To coordinate the specific gravity of affinity and concentration；
The probability value that antibody is selected according to affinity with concentration is sorted from high to low, selects the highest preceding N of probability valuepIt is a
Antibody is as new population particle, and for the optimizing of particle next time, then go to step S3；
S12: the individual adaptive optimal control angle value of each particle and global optimum's fitness value of all particles are updated；
S13: judging whether the maximum number of iterations for reaching immune Chaos particle swarm optimization algorithm, if so, exporting the overall situation of the algorithm
The optimal case that the corresponding particle of adaptive optimal control angle value is distributed as task, otherwise go to step S3.
3. a kind of distributed computing method of smart grid-oriented hybrid network framework according to claim 2, feature
It is,
The method for establishing system model f the following steps are included:
Assuming that by C1、C2、C3......CLRespectively represent L edge calculations node, CcdA cloud computing node is represented,
The communication overhead between any two node is represented,
Wherein,Indicate any two nodes CIAnd CJBetween relationship,It can only be 0 or 1, whenIt is 0
When, indicate two node CIAnd CJBetween there are the task relations of distribution；WhenWhen being 1, two node C are indicatedIAnd CJBetween do not deposit
In the task relations of distribution；Then indicate any two nodes CIAnd CJBetween data transmission delay；
Assuming that the computing capability of each edge calculations node is used respectivelyIt indicates, cloud computing section
The computing capability of point is usedIt indicates, waiting task is indicated with U, and the son for distributing to edge calculations node and cloud computing node is appointed
Business ratio δIAnd δcdIt indicates, data transmission delayBy transmission delayAnd propagation delayIt indicates,
Wherein, K is data frame length,The bandwidth of communication network between two nodes；
Since hybrid network framework uses distributed computing method, a task is distributed into multiple adjacent node parallel processings,
So the processing time of entire task is equal to the processing time of that subtask of longest when the collaboration of all subtasks is handled, therefore,
The model foundation of whole system is as follows:
4. a kind of distributed computing method of smart grid-oriented hybrid network framework according to claim 1 or 2 or 3,
It is characterized in that, cloud computing layer provides reliable data processing and secure storage service for system；Load balancing layer is in edge meter
Task is distributed between operator node and cloud computing node；Edge calculations layer provides calculating and storage service nearby for equipment；Terminal is set
Standby layer provides various supply and distribution services for system；Transport network layer provides efficient, reliability services for share of system information.
5. a kind of distributed computing method of smart grid-oriented hybrid network framework according to claim 4, feature
It is, the transport network layer is made of low-power consumption wide area network LoRa network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810983999.4A CN109218414A (en) | 2018-08-27 | 2018-08-27 | A kind of distributed computing method of smart grid-oriented hybrid network framework |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810983999.4A CN109218414A (en) | 2018-08-27 | 2018-08-27 | A kind of distributed computing method of smart grid-oriented hybrid network framework |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109218414A true CN109218414A (en) | 2019-01-15 |
Family
ID=64985872
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810983999.4A Pending CN109218414A (en) | 2018-08-27 | 2018-08-27 | A kind of distributed computing method of smart grid-oriented hybrid network framework |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109218414A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110365753A (en) * | 2019-06-27 | 2019-10-22 | 北京邮电大学 | Internet of Things service low time delay load allocation method and device based on edge calculations |
| CN110717300A (en) * | 2019-09-27 | 2020-01-21 | 云南电网有限责任公司 | Edge calculation task allocation method for real-time online monitoring service of power internet of things |
| CN110839220A (en) * | 2019-10-28 | 2020-02-25 | 无锡职业技术学院 | Distributed computing method and system based on wireless ad hoc network |
| CN111131421A (en) * | 2019-12-13 | 2020-05-08 | 中国科学院计算机网络信息中心 | Method for interconnection and intercommunication of industrial internet field big data and cloud information |
| CN111901145A (en) * | 2020-06-23 | 2020-11-06 | 国网江苏省电力有限公司南京供电分公司 | Power Internet of things heterogeneous shared resource allocation system and method |
| CN111988787A (en) * | 2020-07-27 | 2020-11-24 | 山东师范大学 | Method and system for selecting network access and service placement positions of tasks |
| CN112084026A (en) * | 2020-09-02 | 2020-12-15 | 国网河北省电力有限公司石家庄供电分公司 | Low-energy-consumption edge computing resource deployment system and method based on particle swarm |
| CN112256413A (en) * | 2020-10-16 | 2021-01-22 | 国网电子商务有限公司 | Scheduling method and device for edge computing task based on Internet of things |
| CN112954022A (en) * | 2020-12-29 | 2021-06-11 | 广东电网有限责任公司电力科学研究院 | Multi-concurrency real-time communication method and device based on intelligent substation |
| CN114900518A (en) * | 2022-04-02 | 2022-08-12 | 中国光大银行股份有限公司 | Task allocation method, device, medium and electronic equipment for directed distributed network |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102118332A (en) * | 2011-04-14 | 2011-07-06 | 南京信息工程大学 | Orthogonal wavelet blind equalization method based on immune clone particle swarm optimization |
| US20120130929A1 (en) * | 2010-11-24 | 2012-05-24 | International Business Machines Corporation | Controlling quarantining and biasing in cataclysms for optimization simulations |
| CN103903054A (en) * | 2014-04-23 | 2014-07-02 | 武汉大学 | Self-learning transit particle swarm artificial intelligence algorithm |
| CN104679966A (en) * | 2015-03-26 | 2015-06-03 | 孙凌宇 | Empowerment hypergraph optimized partitioning method based on multilayer method and discrete particle swarm |
| CN105186556A (en) * | 2015-08-20 | 2015-12-23 | 国家电网公司 | Large photovoltaic power station reactive optimization method based on improved immune particle swarm optimization algorithm |
| CN105430706A (en) * | 2015-11-03 | 2016-03-23 | 国网江西省电力科学研究院 | WSN (Wireless Sensor Networks) routing optimization method based on improved PSO (particle swarm optimization) |
| CN107578128A (en) * | 2017-08-31 | 2018-01-12 | 南京理工大学 | Across level distribution network planing method based on immunity particle cluster algorithm |
-
2018
- 2018-08-27 CN CN201810983999.4A patent/CN109218414A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120130929A1 (en) * | 2010-11-24 | 2012-05-24 | International Business Machines Corporation | Controlling quarantining and biasing in cataclysms for optimization simulations |
| CN102118332A (en) * | 2011-04-14 | 2011-07-06 | 南京信息工程大学 | Orthogonal wavelet blind equalization method based on immune clone particle swarm optimization |
| CN103903054A (en) * | 2014-04-23 | 2014-07-02 | 武汉大学 | Self-learning transit particle swarm artificial intelligence algorithm |
| CN104679966A (en) * | 2015-03-26 | 2015-06-03 | 孙凌宇 | Empowerment hypergraph optimized partitioning method based on multilayer method and discrete particle swarm |
| CN105186556A (en) * | 2015-08-20 | 2015-12-23 | 国家电网公司 | Large photovoltaic power station reactive optimization method based on improved immune particle swarm optimization algorithm |
| CN105430706A (en) * | 2015-11-03 | 2016-03-23 | 国网江西省电力科学研究院 | WSN (Wireless Sensor Networks) routing optimization method based on improved PSO (particle swarm optimization) |
| CN107578128A (en) * | 2017-08-31 | 2018-01-12 | 南京理工大学 | Across level distribution network planing method based on immunity particle cluster algorithm |
Non-Patent Citations (1)
| Title |
|---|
| SHIYUAN NING,QUANBO GE,HAOYU JIANG: "Research on Distributed Computing Method for Coordinated Cooperation of Distributed Energy and Multi-devices", 《2018 33RD YOUTH ACADEMIC ANNUAL CONFERENCE OF CHINESE ASSOCIATION OF AUTOMATION (YAC)》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110365753A (en) * | 2019-06-27 | 2019-10-22 | 北京邮电大学 | Internet of Things service low time delay load allocation method and device based on edge calculations |
| CN110717300A (en) * | 2019-09-27 | 2020-01-21 | 云南电网有限责任公司 | Edge calculation task allocation method for real-time online monitoring service of power internet of things |
| CN110717300B (en) * | 2019-09-27 | 2022-10-21 | 云南电网有限责任公司 | Edge calculation task allocation method for real-time online monitoring service of power internet of things |
| CN110839220A (en) * | 2019-10-28 | 2020-02-25 | 无锡职业技术学院 | Distributed computing method and system based on wireless ad hoc network |
| CN110839220B (en) * | 2019-10-28 | 2022-12-20 | 无锡职业技术学院 | Distributed computing method based on wireless ad hoc network |
| CN111131421B (en) * | 2019-12-13 | 2022-07-29 | 中国科学院计算机网络信息中心 | Method for interconnection and intercommunication of industrial internet field big data and cloud information |
| CN111131421A (en) * | 2019-12-13 | 2020-05-08 | 中国科学院计算机网络信息中心 | Method for interconnection and intercommunication of industrial internet field big data and cloud information |
| CN111901145A (en) * | 2020-06-23 | 2020-11-06 | 国网江苏省电力有限公司南京供电分公司 | Power Internet of things heterogeneous shared resource allocation system and method |
| CN111988787A (en) * | 2020-07-27 | 2020-11-24 | 山东师范大学 | Method and system for selecting network access and service placement positions of tasks |
| CN111988787B (en) * | 2020-07-27 | 2023-04-28 | 山东师范大学 | Task network access and service placement position selection method and system |
| CN112084026A (en) * | 2020-09-02 | 2020-12-15 | 国网河北省电力有限公司石家庄供电分公司 | Low-energy-consumption edge computing resource deployment system and method based on particle swarm |
| CN112084026B (en) * | 2020-09-02 | 2024-05-17 | 国网河北省电力有限公司石家庄供电分公司 | Particle swarm-based low-energy-consumption edge computing resource deployment system and method |
| CN112256413A (en) * | 2020-10-16 | 2021-01-22 | 国网电子商务有限公司 | Scheduling method and device for edge computing task based on Internet of things |
| CN112954022A (en) * | 2020-12-29 | 2021-06-11 | 广东电网有限责任公司电力科学研究院 | Multi-concurrency real-time communication method and device based on intelligent substation |
| CN114900518A (en) * | 2022-04-02 | 2022-08-12 | 中国光大银行股份有限公司 | Task allocation method, device, medium and electronic equipment for directed distributed network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109218414A (en) | A kind of distributed computing method of smart grid-oriented hybrid network framework | |
| Yu et al. | Distributed real-time energy management in data center microgrids | |
| Chen et al. | Efficient multi-vehicle task offloading for mobile edge computing in 6g networks | |
| Taher et al. | New approach for optimal UPFC placement using hybrid immune algorithm in electric power systems | |
| Wang et al. | Exploring smart grid and data center interactions for electric power load balancing | |
| Liu et al. | Economical and balanced energy usage in the smart home infrastructure: A tutorial and new results | |
| Yu et al. | Distributed online energy management for data centers and electric vehicles in smart grid | |
| Cong et al. | EdgeGO: A mobile resource-sharing framework for 6G edge computing in massive IoT systems | |
| Yasmeen et al. | Efficient resource provisioning for smart buildings utilizing fog and cloud based environment | |
| Zhou et al. | A new framework for peer-to-peer energy sharing and coordination in the energy internet | |
| Bakken et al. | Grip-grids with intelligent periphery: Control architectures for grid2050 π | |
| Li et al. | Towards dynamic pricing-based collaborative optimizations for green data centers | |
| Wang et al. | Robust smart-grid-powered cooperative multipoint systems | |
| CN109950907A (en) | The dispatching method and system of alternating current-direct current mixing power distribution network containing electric power electric transformer | |
| Li et al. | Joint optimization of bandwidth for provider and delay for user in software defined data centers | |
| Loganathan et al. | Smart energy management systems: a literature review | |
| Du et al. | Efficient forecasting scheme and optimal delivery approach of energy for the energy Internet | |
| Liu et al. | A multi-user computation offloading algorithm based on game theory in mobile cloud computing | |
| Fatima et al. | A brief review on smart grid residential network schemes | |
| Khalid et al. | An evolutionary approach to optimize data center profit in smart grid environment | |
| Zhang et al. | Energy scheduling for networked microgrids with co-generation and energy storage | |
| US20210021130A1 (en) | Systems and methods for distributed hierarchical artificial intelligence in smart grids | |
| Mahmud et al. | A distributed framework for carbon and cost aware geographical job scheduling in a hybrid data center infrastructure | |
| CN116937584A (en) | Micro-grid group multi-objective optimization solving method and device based on improved wolf algorithm | |
| Ke | Bipartite Graph based Edge Server Placement Algorithm in Mobile Edge Computing |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190115 |
|
| RJ01 | Rejection of invention patent application after publication |