CN111314123B - Time delay and energy consumption-oriented power Internet of things work load distribution method - Google Patents
Time delay and energy consumption-oriented power Internet of things work load distribution method Download PDFInfo
- Publication number
- CN111314123B CN111314123B CN202010079874.6A CN202010079874A CN111314123B CN 111314123 B CN111314123 B CN 111314123B CN 202010079874 A CN202010079874 A CN 202010079874A CN 111314123 B CN111314123 B CN 111314123B
- Authority
- CN
- China
- Prior art keywords
- energy consumption
- time delay
- aerial vehicle
- unmanned aerial
- vehicle terminal
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000005265 energy consumption Methods 0.000 claims abstract description 134
- 238000004364 calculation method Methods 0.000 claims abstract description 38
- 238000005457 optimization Methods 0.000 claims abstract description 25
- 239000013598 vector Substances 0.000 claims description 42
- 238000012545 processing Methods 0.000 claims description 22
- 230000005540 biological transmission Effects 0.000 claims description 20
- 239000011159 matrix material Substances 0.000 claims description 19
- 230000033001 locomotion Effects 0.000 claims description 17
- 230000003321 amplification Effects 0.000 claims description 8
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 8
- 230000002068 genetic effect Effects 0.000 claims description 6
- 238000013468 resource allocation Methods 0.000 claims description 6
- 238000007689 inspection Methods 0.000 claims description 4
- 230000001934 delay Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 229910052731 fluorine Inorganic materials 0.000 claims description 2
- 125000001153 fluoro group Chemical group F* 0.000 claims description 2
- 230000002035 prolonged effect Effects 0.000 abstract description 2
- 230000004083 survival effect Effects 0.000 abstract description 2
- 230000005012 migration Effects 0.000 description 14
- 238000013508 migration Methods 0.000 description 14
- 230000008569 process Effects 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 6
- 230000007547 defect Effects 0.000 description 5
- 206010042135 Stomatitis necrotising Diseases 0.000 description 3
- 201000008585 noma Diseases 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000001413 cellular effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000004931 aggregating effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
- H04L41/0833—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability for reduction of network energy consumption
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0803—Configuration setting
- H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5003—Managing SLA; Interaction between SLA and QoS
-
- 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
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
-
- 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
- H04L67/1004—Server selection for load balancing
- H04L67/101—Server selection for load balancing based on network conditions
-
- 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
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
- H04W28/24—Negotiating SLA [Service Level Agreement]; Negotiating QoS [Quality of Service]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/14—Relay systems
- H04B7/15—Active relay systems
- H04B7/185—Space-based or airborne stations; Stations for satellite systems
- H04B7/18502—Airborne stations
- H04B7/18506—Communications with or from aircraft, i.e. aeronautical mobile service
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/02—Hierarchically pre-organised networks, e.g. paging networks, cellular networks, WLAN [Wireless Local Area Network] or WLL [Wireless Local Loop]
- H04W84/04—Large scale networks; Deep hierarchical networks
- H04W84/06—Airborne or Satellite Networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The application discloses a time delay and energy consumption-oriented power Internet of things work load distribution method, which comprises the following steps: constructing a power internet of things work load distribution model based on edge calculation; on the basis of the constructed power Internet of things work load distribution model, a multi-objective optimization function of power Internet of things work load distribution is established by taking the time delay and the energy consumption of the unmanned aerial vehicle terminal UE as a common optimization objective; and solving the established multi-objective optimization function by improving the MOEA/D algorithm. The application provides a workload distribution method based on an improved MOEA/D algorithm, and the capability of the algorithm for searching an optimal solution is improved; the maximum fitness increment is used in the neighborhood updating strategy to update the bad individuals in the neighborhood, so that the survival time of the excellent individuals is prolonged, the final result is closer to the optimal workload distribution mode, the energy consumption under the same time delay index is smaller, and the time delay under the same energy consumption index is smaller.
Description
Technical Field
The invention belongs to the technical field of power internet of things, relates to a time delay and energy consumption-oriented power internet of things work load distribution technology, and particularly relates to a time delay and energy consumption-oriented power internet of things work load distribution method based on edge calculation.
Background
The power internet of things is an important application practice of the internet of things technology in the power industry. With the continuous development of image recognition technology and emerging communication technology, the specialized inspection work in transmission lines and transformer substations has been carried out to many rotor unmanned aerial vehicle terminal UE in the electric power thing networking, through patrolling and examining equipment closely infrared and visible light, the defect position is accurately confirmed, various trouble hidden dangers that ground is difficult to discover equipment are discovered. Unmanned aerial vehicle terminal UE carries out the high altitude and closely observes, acquires the clear image data of framework, lightning rod, in time discovers the abnormal conditions of equipment operation through contrastive analysis, provides the powerful guarantee for electric power thing networking safety and stability moves.
The edge computing technology can remarkably reduce the request time delay of the unmanned aerial vehicle terminal UE in the power internet of things by sinking the cloud computing capability to the edge side of the network, and meets the higher requirement of the service QoS of the unmanned aerial vehicle terminal UE. In the electric power internet of things based on edge calculation, an unmanned aerial vehicle terminal UE is accessed into the electric power internet of things, and a calculation task is distributed to different edge nodes EN for distributed processing. Due to the hardware technology in the current stage, the battery energy of the unmanned aerial vehicle terminal UE is limited, and the strict low-delay requirement of the unmanned aerial vehicle terminal UE cannot be met within the limited endurance time. On the one hand, the drone terminal UE requests are intended to be processed quickly, and on the other hand, the energy consumed by the drone terminal UE is as small as possible in order to be able to maintain a longer cruising time of the drone terminal UE. How to balance the optimization targets of time delay and energy consumption and find a compromise solution between the time delay and the energy consumption is a difficult point for the unmanned aerial vehicle terminal UE to patrol and examine the workload distribution of the service.
In order to solve the development situation of the prior art, the existing papers and patents are searched, compared and analyzed, and the following technical information with high relevance to the invention is screened out:
the technical scheme 1: a patent of "an optimization method of energy consumption and time delay and minimization in edge calculation" with publication number CN109600178A, which relates to an optimization method of energy consumption and time delay and minimization in edge calculation, and is mainly completed by four steps: firstly, calculating the optimal power of cellular users in all NOMA groups, and randomly selecting N NOMA groups; secondly, randomly matching N NOMA groups and D2D pairs one by using an edge server, and calculating a utility function; thirdly, performing pre-exchange operation by using an edge server; fourthly, calculating a utility function by utilizing the edge server, and executing the exchange operation if the exchange condition is met.
The technical scheme 1 achieves the purpose of calculating energy consumption and time delay and minimizing of cellular and D2D users in the migration process, and has the defects that the calculation task needs to be migrated for multiple times between edge servers, additional time delay needs to be consumed in the task migration process, and the time delay sensitive task cannot meet the QoS requirement.
The technical scheme 2 is as follows: patent of a mobile terminal computation migration method for optimizing time delay and energy efficiency with publication number CN108376099A, relating to a mobile terminal computation migration method for optimizing time delay and energy efficiency, which is mainly completed by three steps: firstly, establishing a calculation migration model of a wireless terminal; secondly, constructing a migration cost function on the basis of the model; thirdly, with the reduction of time delay and the reduction of energy consumption as constraint conditions, computing migration is reasonably implemented by analyzing the requirements of application programs, the computing power of the mobile terminal and the wireless channel rate, and the aim of comprehensively optimizing the operation time delay and the energy consumption of the mobile terminal is achieved.
The technical scheme 2 is suitable for computing migration of LTE network application, an application program of the mobile terminal is decomposed into a plurality of computing components, computing parameters are obtained according to relevance among the components, a migration cost function for judging migration cost of the computing components by a user is constructed, computing migration conditions for reducing time delay and improving energy efficiency of each computing component are obtained, the migration conditions are converted into an optimization problem, and migration decision is implemented according to a solving result. The method has the disadvantages that the decomposition and modeling of the computing components require a certain computing power of the terminal, the intelligent requirement on the terminal is higher, and an accurate computing migration model cannot be established for the acquisition terminal with lower intelligent degree.
The technical scheme 3 is as follows: a patent of 'a method and a device for D2D task allocation based on moving edge calculation' with publication number CN108319502A, which relates to a method for D2D task allocation based on moving edge calculation, and is mainly completed through three steps: firstly, generating a task allocation scheme set according to information of each network device in a heterogeneous network where a target device for generating a task is located; secondly, acquiring user requirements of the target equipment, wherein the user requirements comprise an energy consumption index and a time delay index; thirdly, generating an energy consumption weight and a time delay weight according to the corresponding relation of the energy consumption index, the time delay index and a preset index weight; and fourthly, selecting an optimal task allocation scheme in the task allocation scheme set according to the energy consumption weight and the time delay weight.
The technical scheme 3 can intelligently adjust the task allocation tendency, and has the defects that the task allocation scheme excessively depends on an initially generated task allocation scheme set, the problem of weight of energy consumption indexes and time delay indexes is difficult to solve in the scene of dynamic change of terminals and tasks, and the limitation for obtaining the optimal task allocation scheme is large.
Disclosure of Invention
In order to solve the defects in the prior art, the application provides a time delay and energy consumption-oriented power internet of things work load distribution method, cross distribution indexes of genetic operators are dynamically modified according to the population evolution process of an iterative algorithm, bad individuals in a neighborhood are updated by using a maximum fitness increment index in a neighborhood updating strategy, so that individual distribution is improved, the iteration speed and result approximation degree are improved, and the final work load distribution scheme meets the requirement that time delay and energy consumption of unmanned aerial vehicle terminal UE are minimum.
In order to achieve the above objective, the following technical solutions are adopted in the present application:
a time delay and energy consumption oriented power Internet of things work load distribution method comprises the following steps:
step 1: aiming at the power internet of things, the power internet of things is accessed through an unmanned aerial vehicle terminal UE, a calculation task is distributed to different edge nodes EN for distributed processing in an unmanned aerial vehicle terminal UE inspection scene, and a power internet of things working load distribution model based on edge calculation is constructed;
step 2: based on the power Internet of things work load distribution model constructed in the step 1, a multi-objective optimization function of power Internet of things work load distribution is established by taking the time delay and the energy consumption of the unmanned aerial vehicle terminal UE as a common optimization objective;
and step 3: and (3) solving the multi-objective optimization function established in the step (2) by improving the MOEA/D algorithm to obtain a power Internet of things work load distribution scheme.
The invention further comprises the following preferred embodiments:
preferably, the time delay of the drone terminal UE includes a network time delay of the task transmitted from the drone terminal UE to the edge node EN and a calculation time delay of the task processed on the edge node EN;
the energy consumption of the unmanned aerial vehicle terminal UE comprises the motion energy consumption when the unmanned aerial vehicle terminal UE normally executes a task to cruise and the transmission energy consumption when the unmanned aerial vehicle terminal UE transmits data to the edge node EN.
Preferably, the power internet of things workload distribution model constructed in step 1 considers the edge node EN set in one region asThe unmanned aerial vehicle terminal UE is set asUnmanned aerial vehicle terminal UE j The upper task block request is set asWherein j is the number of the unmanned aerial vehicle terminal UE, and the unmanned aerial vehicle terminal UE j The last k task block request represents w by a vector jk =[b jk ,ω jk ];
Wherein, b jk Indicating the amount of data that needs to be transmitted; omega jk Representing the load size of the task, i.e. the number of instructions that the CPU needs to execute.
Preferably, the time delay of the UE in step 2 is the sum of the time delays of all task blocks in the UE, and when one task block in the UE is an indivisible task and each task block requests to be allocated to only one edge node EN for processing, the calculation formula of the time delay D of the UE is as follows:
wherein, d j For a single drone terminal UE j Time delay of (2); d jk For the delay of one task block request in the drone terminal UE,for the network delay from the drone terminal UE to the edge node EN,requesting a computation delay on an edge node EN for a task block;
when the distribution mode of the workload satisfies the formulas (4) and (5),andexpressed as (6) and (7), respectively:
x jk =i (4)
y jk =l (5)
wherein x is jk = i denotes task Block request w jk To edge nodes EN i ,y jk = l means at edge node EN i Inner, task block request w jk Assigned to a VM l Carrying out treatment;
wherein r is ij Is an unmanned aerial vehicle terminal UE j To the edge node EN i C is the propagation speed of the radio channel;requesting w for a task block jk The calculated time delay at the edge node EN,is w jk At edge node EN i Virtual machine VM l Time delay of processing, w jk Indicating unmanned aerial vehicle terminal UE j The last kth task block request, q il (w jk ) Indicating w in a task block request sequence jk Sum of all task block requests, ω, pending ahead jk Representing task Block requests w jk Load size of v il Representing edge nodes EN i Middle virtual machine VM l The processing rate of the CPU of (1); omega j'k' Load size, x, requested for a task block in a sequence of task block requests j'k' = i denotes that this task block request is allocated to the edge node EN i 。
Preferably, in step 2, the energy consumption of the UE is the sum of the energy consumptions of all UEs, and the calculation formula of the energy consumption E of the UE is:
wherein e is j For a single drone terminal UE j The energy consumption of (2) is reduced,for a single drone terminal UE j The energy consumption of the transmission of (2),for a single drone terminal UE j Energy consumption of movement of b jk Indicating the amount of data to be transmitted, e tc Represents a parameter of the energy consumption of the transmitting circuit (transmitting circuit), e pa Represents a power amplification (power amplification) energy consumption parameter, and r is a transmission distance; e.g. of the type base Representing a basic kinetic energy consumption parameter, alpha j Is a basic motion energy consumption adjustment factor, independent of the workload distribution mode of the unmanned aerial vehicle terminal UE, e add Representing an additional kinetic energy consumption parameter, beta j Adjusting factors for additional motion energy consumption;and distributing Euclidean distance between vectors for optimizing the working load of the front unmanned aerial vehicle terminal UE and the rear unmanned aerial vehicle terminal UE.
Preferably, the multi-objective optimization function of the power internet of things workload distribution established in step 2 is as follows:
0≤ν il ≤ν i (16)
d (x, v) represents the time delay of the unmanned aerial vehicle terminal UE; e (x) represents the energy consumption of the unmanned aerial vehicle terminal UE;
x=(x jk ) JK×1 a task allocation matrix X = (X) for mapping a task block request in the drone terminal UE to the edge node EN jk ) J×K Is in vector form, matrix element x jk Representing task Block requests w jk The number of the edge node EN assigned;
v=(ν il ) IL×1 and allocating a matrix V = (V) to the virtual machine VM in the edge node EN il ) I×L In the form of column vectors of matrix element v il Representing edge nodes EN i Middle virtual machine VM l The CPU processing rate of (1);
ν i representing the processing rate of the CPU in the edge node EN;
d j for a single drone terminal UE j The time delay of (2) is the sum of the network time delay and the calculation time delay;for unmanned aerial vehicle terminal UE j A delay constraint threshold of (c);
e j for a single drone terminal UE j The energy consumption of (2) is the sum of the motion energy consumption and the transmission energy consumption;for unmanned aerial vehicle terminal UE j The energy consumption constraint threshold of (c).
Preferably, the step 3 of solving the multi-objective optimization function established in the step 2 by improving the MOEA/D algorithm comprises the following steps:
step 301: set of input edge nodes ENSet of unmanned aerial vehicle terminals UEEach unmanned aerial vehicle terminal UE j Set of upper task block requestsTask Block request w jk =[b jk ,ω jk ]Virtual machine VM distribution matrix V, time delay constraint thresholdEnergy consumption constraint thresholdUE j To EN i Is a physical distance r ij Energy consumption parameter e of transmitting circuit tc Power amplification energy consumption parameter e pa Basic kinetic energy consumption parameter e base Adjustment factor alpha of basic kinetic energy consumption j And an additional kinetic energy consumption parameter e add ;
Step 302: initializing the computing resource allocation, the direction vector and the task allocation of an edge node EN;
step 303: calculating the distance between any two direction vectors;
step 304: selecting the nearest T direction vectors to be placed in the neighborhood;
step 305: reference point Z for initialization * ;
Step 306: in the range of the iteration times, each individual neighborhood is updated in an iteration mode;
step 307: and outputting the task allocation matrix X.
Preferably, the step 306 of updating each individual neighborhood is implemented as follows:
step (1): calculating a distribution cross index eta, wherein the calculation formula is as follows:
wherein s represents the number of iterations;
step (2): calculating a distribution factor rho by the following calculation formula:
wherein u is a random number within (0, 1).
And (3): calculating the offspring individuals, wherein the calculation formula of the offspring individuals is as follows:
wherein, rho is a distribution factor,andis a parent individual and is a new parent individual,andis an offspring individual.
And (4): updating the reference point;
and (5): judging the number of the neighborhood renewable individuals, and selecting a better individual to enter the next generation according to the optimization objective function to update the neighborhood when the number of the neighborhood renewable individuals is more than zero.
Preferably, in step (5), the neighborhood is updated by using a 2-norm constrained direction vector and a Tchebycheff method to solve the nonlinear relation in the objective function, and the specific steps are as follows:
step (5-1): calculating an objective function value:
let lambda 1 ,λ 2 ,…,λ N Is a set of uniformly distributed directional vectors, and z * Is a reference point, according to the Tchebycheff method, for a parameter ofThe objective function of (1) is:
wherein x is a task allocation vector, v is a resource allocation vector, g 2tch (. Cndot.) is a function of an objective,andrespectively being the nth direction vector lambda n Time delay and energy consumption component ofAnd isz * Is an ideal reference vector and is a vector of the reference,is a component of the delay direction, satisfies For consuming energyComponent of direction, satisfyD and E are respectively time delay and energy consumption calculated according to x;
step (5-2): update individuals in the neighborhood:
the individual serial number pi of the maximum fitness increment in the neighborhood is obtained through the following formula, and x is updated by y π :
Wherein y is an offspring individual generated after genetic operator updating and represents an optimal individual in the subproblem; x is a radical of a fluorine atom π Is the calculated individual with the largest fitness increment in the neighborhood.
The beneficial effect that this application reached:
1. in order to comprehensively consider the time delay and the energy consumption index in the work load distribution process, the work load distribution method based on the improved MOEA/D algorithm is provided, in the genetic evolution process, cross distribution factors are dynamically adjusted according to the iteration process, the individual distribution capability is improved, the defect that the original genetic operator is easy to fall into the local optimal solution is overcome, the capability of the algorithm for searching the optimal solution is improved, and therefore the time delay and the energy consumption of the unmanned aerial vehicle terminal UE are reduced in the work load distribution mode.
2. The maximum fitness increment is used in the neighborhood updating strategy to update the bad individuals in the neighborhood, so that the survival time of the excellent individuals is prolonged, the overall level of the population is improved instead of the level of a single individual, the final result is closer to the optimal workload distribution mode, the energy consumption under the same time delay index is smaller, and the time delay under the same energy consumption index is smaller.
Drawings
Fig. 1 is a schematic flow chart of a time delay and energy consumption oriented power internet of things workload distribution method according to the present application;
FIG. 2 is a schematic structural diagram of a workload distribution model established in an embodiment of the present application;
FIG. 3 is a flow chart of workload distribution based on the improved MOEA/D algorithm in the embodiment of the present application;
FIG. 4 is a comparison graph of function values of different population sizes and iteration times in the improved MOEA/D algorithm in the embodiment of the present application;
FIG. 5 is a distribution diagram of time delay after a plurality of experimental results in the embodiment of the present application;
FIG. 6 is a graph showing the distribution of energy consumption after a plurality of experimental results in the examples of the present application.
Detailed Description
The present application is further described below with reference to the accompanying drawings. The following examples are only used to illustrate the technical solutions of the present invention more clearly, and the protection scope of the present application is not limited thereby.
As shown in fig. 1, the method for distributing work loads of the power internet of things facing to time delay and energy consumption in the application comprises the following steps:
step 1: aiming at the power internet of things, the power internet of things is accessed through an unmanned aerial vehicle terminal UE, a calculation task is distributed to different edge nodes EN for distributed processing in an unmanned aerial vehicle terminal UE inspection scene, and a power internet of things working load distribution model based on edge calculation is constructed;
step 2: on the basis of the power Internet of things work load distribution model constructed in the step 1, a multi-objective optimization function of power Internet of things work load distribution is established by taking the time delay and the energy consumption of the unmanned aerial vehicle terminal UE as a common optimization target;
and step 3: and (3) solving the multi-objective optimization function established in the step (2) by improving the MOEA/D algorithm to obtain a power Internet of things work load distribution scheme.
In the embodiment of the application, the time delay of the unmanned aerial vehicle terminal UE comprises network time delay of the task transmitted to the edge node EN by the unmanned aerial vehicle terminal UE and calculation time delay of the task processed on the edge node EN;
the energy consumption of the unmanned aerial vehicle terminal UE comprises the motion energy consumption when the unmanned aerial vehicle terminal UE normally executes the task to cruise and the transmission energy consumption when the unmanned aerial vehicle terminal UE transmits data to the edge node EN.
As shown in fig. 2, in the power internet of things workload distribution model constructed in step 1, edge nodes EN in a region are considered to be collected intoThe unmanned aerial vehicle terminal UE is set asUnmanned aerial vehicle terminal UE j The upper task block request is set asj is the number of the unmanned terminal UE, and the unmanned terminal UE j The last k task block request represents w by a vector jk =[b jk ,ω jk ];
Wherein, b jk Representing the amount of data that needs to be transmitted; omega jk Representing the load size of the task, i.e. the number of instructions that the CPU needs to execute.
The target function mainly comprises time delay and energy consumption of the unmanned aerial vehicle terminal UE, the time delay D is defined as the time from the generation of a request on the unmanned aerial vehicle terminal UE to the completion of processing on the edge node EN, and the energy consumption E of the unmanned aerial vehicle terminal UE is defined as the energy consumed by the movement and data transmission of the unmanned aerial vehicle terminal UE.
Step 2, when the time delay of the unmanned aerial vehicle terminal UE is the sum of the time delays of all task blocks in the unmanned aerial vehicle terminal UE, and assuming that one task block in the unmanned aerial vehicle terminal UE is an inseparable task and each task block request is only distributed to one edge node EN for processing, the calculation formula of the time delay D of the unmanned aerial vehicle terminal UE is as follows:
wherein, d j For a single drone terminal UE j Time delay of (2); d jk Is unmannedThe time delay of a task block request in the terminal UE comprises the network time delay from the terminal UE to the edge node ENAnd the calculation time delay of the task block request on the edge node ENNamely:
when the distribution mode of the workload satisfies the formulas (4) and (5),andexpressed as (6) and (7), respectively:
x jk =i (4)
y jk =l (5)
wherein x is jk = i denotes task Block request w jk To edge nodes EN i ,y jk = l denotes at edge node EN i Inner, task Block request w jk Allocation to VMs l Carrying out treatment;
wherein r is ij Is an unmanned aerial vehicle terminal UE j To the edge node EN i C is the propagation speed of the radio channel;requesting w for a task block jk The calculated time delay at the edge node EN,is w jk At edge node EN i Virtual machine VM l Time delay of processing, w jk Indicating unmanned aerial vehicle terminal UE j The kth task block request;
the mode that the edge node EN processes the task block request is simplified into an M/M/1 queuing model, if the task reaches the edge node EN, the CPU is idle and processes the task, and if the CPU is processing other tasks, the task enters a task queue. In order to ensure the minimum total service delay, the tasks are sorted according to the load of the tasks in the task queue, and the tasks with small load are processed preferentially. The calculation delay includes the time of queuing and the time of actual processing, namely:
wherein q is il (w jk ) Indicating w in a task block request sequence jk Sum of all task block requests, ω, pending ahead jk Representing task Block requests w jk Load size of v il Representing edge nodes EN i Middle virtual machine VM l The processing rate of the CPU of (1); omega j'k' Load size, x, requested for a task block in a sequence of task block requests j'k' = i denotes that this task block request is allocated to the edge node EN i 。
Step 2, the energy consumption of the unmanned aerial vehicle terminal UE is the sum of the energy consumption of all the unmanned aerial vehicle terminals UE, and a single unmanned aerial vehicle terminal UE j Energy consumption e of j Energy consumption for transmission including data transmissionAnd energy consumption for maintaining self-flying movement
The calculation formula of the energy consumption E of the unmanned aerial vehicle terminal UE is as follows:
the motion energy consumption of the unmanned aerial vehicle terminal UE comprises basic energy consumption and additional energy consumption. Due to the influence of motion states such as steering, hovering, acceleration and deceleration, a part of energy consumption is increased or reduced, and the part of energy consumption changed by the workload distribution mode is called additional energy consumption.
The transmission energy consumption of the unmanned aerial vehicle terminal UE is mainly linearly related to the 2 nd power of the transmission distance r, and after the transmission energy consumption exceeds a certain distance, the transmission energy consumption is in direct proportion to the 4 th power of the distance, and data transmission within the 2 nd power distance is mainly considered in the embodiment of the application.
Thus:
wherein, b jk Indicating the amount of data to be transmitted, e tc Representing a parameter of energy consumption of the transmitting circuit, e pa Representing a power amplification energy consumption parameter, wherein r is a transmission distance; e.g. of the type base Representing a basic kinetic energy consumption parameter, alpha j Is a basic motion energy consumption adjustment factor, independent of the workload distribution mode of the unmanned aerial vehicle terminal UE, e add Representing an additional kinetic energy consumption parameter, beta j Adjusting factors for additional kinetic energy consumption:
wherein,and distributing Euclidean distance between vectors for optimizing the working load of the front unmanned aerial vehicle terminal UE and the rear unmanned aerial vehicle terminal UE.
The sum of the computing power of all VMs in one edge node EN should not be greater than the actual computing power of the edge node EN, i.e. the following constraint is satisfied:
0≤ν il ≤ν i (16)
V=(ν il ) I×L is a VM allocation matrix representing the edge node EN, and the matrix element v il Representing edge nodes EN i Middle virtual machine VM l The CPU processing rate of (1) is developed into a column vector form of v = (v) il ) IL×1 ;ν i Indicating the processing rate of the CPU in the edge node EN.
X=(x jk ) J×K Is a task allocation matrix representing the mapping between the task block request in the unmanned aerial vehicle terminal UE and the edge node EN, and the matrix element x jk Is defined as shown in formula (17), matrix element x jk Representing task Block requests w jk Number assigned to edge node EN, X = (X) jk ) J×K Unfolding the columns into vector form as x = (x) jk ) JK×1 ;
The unmanned aerial vehicle terminal UE has certain QoS requirements as the communication unmanned aerial vehicle terminal UE, and the time delay and the energy consumption cannot be larger than a certain threshold, so that the following constraints are met.
d j For a single drone terminal UE j The time delay of (1) is the sum of the network time delay and the calculation time delay;for unmanned terminal UE j A delay constraint threshold of;
e j for a single drone terminal UE j The energy consumption of (2) is the sum of the motion energy consumption and the transmission energy consumption;for unmanned aerial vehicle terminal UE j The energy consumption constraint threshold of (c).
In summary, the multi-objective optimization function for the power internet of things workload distribution established in step 2 is as follows:
d (x, v) represents the time delay of the unmanned aerial vehicle terminal UE; e (x) represents the energy consumption of the drone terminal UE.
As shown in fig. 3, the step 3 of solving the multi-objective optimization function established in the step 2 by improving the MOEA/D algorithm includes the following steps:
step 301: set of input edge nodes ENSet of unmanned aerial vehicle terminals UEEach unmanned aerial vehicle terminal UE j Set of upper task block requestsTask Block request w jk =[b jk ,ω jk ]Virtual machine VM distribution matrix V, time delay constraint thresholdEnergy consumption constraint thresholdUnmanned aerial vehicle terminal UE j To the edge node EN i Is a physical distance r ij Energy consumption parameter e of transmitting circuit tc Power amplification energy consumption parameter e pa Basic kinetic energy consumption parameter e base Adjustment factor alpha of basic kinetic energy consumption j And an additional kinetic energy consumption parameter e add ;
Step 302: computing resource allocation, direction vector and task allocation of an edge node EN are initialized;
step 303: calculating the distance between any two direction vectors;
step 304: selecting the nearest T direction vectors to be placed in a neighborhood;
step 305: reference point Z of initialization * ;
Step 306: in the range of the iteration times, each individual neighborhood is updated in an iteration mode;
step 307: and outputting the task allocation matrix X.
Step 306, updating each individual neighborhood, the implementation steps are as follows:
step (1): calculating a distribution cross index eta:
in the initial stage of the evolution process, a small distribution cross index eta is used for carrying out dispersion search, which is beneficial to exploring unknown spatial information and keeping the diversity of solutions; along with the evolution, the solution individuals tend to converge, eta is gradually increased, small-range centralized search is carried out, and the convergence speed is increased. The cross-distribution index is calculated by the following formula:
wherein s represents the number of iterations;
step (2): calculating a distribution factor rho, wherein for the convenience of calculating offspring individuals according to parent individuals, a calculation formula of rho is as follows:
wherein u is a random number within (0, 1).
And (3): calculating the offspring individuals, wherein the calculation formula of the offspring individuals is as follows:
wherein, rho is a distribution factor,andis a parent individual and is a new parent individual,andis an offspring individual.
And (4): updating the reference point;
and (5): judging the number of the neighborhood renewable individuals, selecting a better individual to enter the next generation according to the optimization objective function when the number of the neighborhood renewable individuals is more than zero (the better individual is calculated according to a formula (24)), and updating the neighborhood.
In the step (5), the neighborhood is updated by using a direction vector of 2 norm constraint and a Tchebycheff method to solve the nonlinear relation in the objective function, the strategy for updating the neighborhood updates individuals with the maximum fitness in the population all the time, excellent individuals exist all the time, and the overall level of the population is improved instead of single individuals.
The method comprises the following specific steps:
step (5-1): calculating an objective function value:
let lambda be 1 ,λ 2 ,…,λ N Is a set of uniformly distributed directional vectors, and z * Is a reference point, according to the Tchebycheff method, for a parameter ofThe objective function of the sub-problem of (1) is:
wherein x is a task allocation vector, v is a resource allocation vector, g 2tch (. Cndot.) is a function of an objective,andrespectively being the nth direction vector lambda n Time delay and energy consumption component ofAnd isz * Is an ideal reference vector and is a vector of the reference, is a component of the delay direction, satisfies Is a component of the direction of energy consumptionD and E are respectively time delay and energy consumption calculated according to x;
step (5-2): updating individuals in the neighborhood:
the individual serial number pi of the maximum fitness increment in the neighborhood is obtained through the following formula, and x is updated by y π :
Wherein y is the filial generation individuals generated after genetic operator updating and represents the optimal individuals in the subproblem, and x π Is the calculated individual with the largest fitness increment in the neighborhood.
The embodiment of the application verifies the performance of the proposed scheme by using the numerical result. Regard as unmanned aerial vehicle terminal UE with four rotor unmanned aerial vehicle, concrete energy consumption parameter is as follows: the energy consumption of the transmitting circuit is 50nJ/bit, and the energy consumption of the power amplification is 10pJ/bit/m 2 The distance threshold of wireless transmission is 100m, the mobile power consumption is 240W, and the hovering power consumption is 210W.
Fig. 4 is a functional value showing the correspondence between different population sizes and the number of iterations in the improved MOEA/D algorithm, where the functional value is the result of aggregating the time delay and the energy consumption according to different weights. As shown in fig. 4, as the population size increases and the number of iterations increases, the delay and power consumption gradually decrease, and the decrease becomes smaller and smaller. When the population size is larger than 160 and the iteration number is larger than 2000, the function value area is stable, and the optimal solution of the workload distribution is obtained.
As shown in fig. 5, the time delay values of the UE of the drone terminal obtained by the workload distribution scheme obtained in the present application are stably distributed between 220 and 240ms, and are distributed more intensively, which indicates that the convergence and stability of the algorithm are better.
As shown in fig. 6, the energy consumption values of the UE of the unmanned aerial vehicle terminal obtained by the workload distribution scheme obtained in the present application are stably distributed between 430 kJ and 460kJ, the distribution is concentrated, and most of the data is located between the lower quartile and the upper quartile.
The present applicant has described and illustrated embodiments of the present invention in detail with reference to the accompanying drawings, but it should be understood by those skilled in the art that the above embodiments are merely preferred embodiments of the present invention, and the detailed description is only for the purpose of helping the reader to better understand the spirit of the present invention, and not for limiting the scope of the present invention, and on the contrary, any improvement or modification made based on the spirit of the present invention should fall within the scope of the present invention.
Claims (6)
1. A time delay and energy consumption-oriented power Internet of things work load distribution method is characterized by comprising the following steps:
the method comprises the following steps:
step 1: aiming at the power internet of things, the power internet of things is accessed through an unmanned aerial vehicle terminal UE, a calculation task is distributed to different edge nodes EN for distributed processing in an unmanned aerial vehicle terminal UE inspection scene, and a power internet of things working load distribution model based on edge calculation is constructed;
step 2: on the basis of the power Internet of things work load distribution model constructed in the step 1, a multi-objective optimization function of power Internet of things work load distribution is established by taking the time delay and the energy consumption of the unmanned aerial vehicle terminal UE as a common optimization target;
and step 3: solving the multi-objective optimization function established in the step 2 by improving an MOEA/D algorithm to obtain a power Internet of things work load distribution scheme;
and 3, solving the multi-objective optimization function established in the step 2 by improving the MOEA/D algorithm, wherein the method comprises the following steps:
step 301: set of input edge nodes ENSet of unmanned aerial vehicle terminals UEEach nobodyMachine terminal UE j Set of upper task block requestsTask Block request w jk =[b jk ,ω jk ]Virtual machine VM distribution matrix V, time delay constraint thresholdEnergy consumption constraint thresholdUE j To EN i Is a physical distance r ij Energy consumption parameter e of the transmitting circuit tc Power amplification energy consumption parameter e pa Basic kinetic energy consumption parameter e base Adjustment factor alpha of basic kinetic energy consumption j And an additional kinetic energy consumption parameter e add ;
Step 302: computing resource allocation, direction vector and task allocation of an edge node EN are initialized;
step 303: calculating the distance between any two direction vectors;
step 304: selecting the nearest T direction vectors to be placed in the neighborhood;
step 305: reference point Z for initialization * ;
Step 306: in the range of the iteration times, each individual neighborhood is updated in an iteration mode;
step 307: outputting a task allocation matrix X;
step 306, updating each individual neighborhood, the implementation steps are as follows:
step (1): calculating a distribution cross index eta, wherein the calculation formula is as follows:
wherein s represents the number of iterations;
step (2): calculating a distribution factor rho by the following calculation formula:
wherein u is a random number within (0, 1);
and (3): calculating the offspring individuals, wherein the calculation formula of the offspring individuals is as follows:
wherein, rho is a distribution factor,andis a parent individual and is a new parent individual,andis an offspring individual;
and (4): updating the reference point;
and (5): judging the number of the neighborhood updatable individuals, and selecting a better individual to enter the next generation according to the optimization objective function to update the neighborhood when the number of the neighborhood updatable individuals is more than zero;
in the step (5), updating the neighborhood by using a direction vector of 2 norm constraint and a Tchebycheff method to solve the nonlinear relation in the objective function, and the specific steps are as follows:
step (5-1): calculating an objective function value:
let lambda be 1 ,λ 2 ,…,λ N Is a set of uniformly distributed directional vectors, and z * Is a reference point, according to the Tchebycheff method, for a parameter ofThe objective function of (1) is:
wherein x is a task allocation vector, v is a resource allocation vector, g 2tch (. Cndot.) is a function of an objective,andrespectively being the nth direction vector lambda n Time delay and energy consumption component ofAnd isz * Is an ideal reference vector and is a vector of the reference,is a component of the delay direction Is a component of the direction of energy consumptionD and E are respectively time delay and energy consumption calculated according to x;
step (5-2): updating individuals in the neighborhood:
the individual serial number pi of the maximum fitness increment in the neighborhood is obtained through the following formula, and x is updated by y π :
Wherein y is an offspring individual generated after genetic operator updating and represents an optimal individual in the subproblem; x is a radical of a fluorine atom π Is the individual with the largest fitness increment in the calculated neighborhood.
2. The time delay and energy consumption oriented power internet of things workload distribution method according to claim 1, characterized in that:
the time delay of the unmanned aerial vehicle terminal UE comprises network time delay of the task transmitted to the edge node EN by the unmanned aerial vehicle terminal UE and calculation time delay of the task processed on the edge node EN;
the energy consumption of the unmanned aerial vehicle terminal UE comprises the motion energy consumption when the unmanned aerial vehicle terminal UE normally executes a task to cruise and the transmission energy consumption when the unmanned aerial vehicle terminal UE transmits data to the edge node EN.
3. The time delay and energy consumption oriented power internet of things workload distribution method according to claim 2, characterized in that:
the power Internet of things work load distribution model constructed in the step 1 considers the fact that edge nodes EN in one region are collected intoThe unmanned aerial vehicle terminal UE is set asUnmanned aerial vehicle terminal UE j The upper task block request is set asj is the serial number of the unmanned plane terminal UE j The last k task block request represents w by a vector jk =[b jk ,ω jk ];
Wherein, b jk Representing the amount of data that needs to be transmitted; omega jk Representing the load size of the task, i.e. the number of instructions that the CPU needs to execute.
4. The time delay and energy consumption oriented power internet of things workload distribution method according to claim 3, characterized in that:
step 2, the time delay of the unmanned aerial vehicle terminal UE is the sum of the time delays of all task blocks in the unmanned aerial vehicle terminal UE, when one task block in the unmanned aerial vehicle terminal UE is an inseparable task and each task block request is only distributed to one edge node EN for processing, the calculation formula of the time delay D of the unmanned aerial vehicle terminal UE is as follows:
wherein d is j For a single drone terminal UE j Time delay of (2); d jk For the delay of one task block request in the drone terminal UE,for the network delay from the drone terminal UE to the edge node EN,requesting a computation delay on an edge node EN for a task block;
when the distribution mode of the workload satisfies the formulas (4) and (5),andrespectively expressed as (6) and (7):
x jk =i (4)
y jk =l (5)
wherein x is jk = i denotes task Block request w jk To edge nodes EN i ,y jk = l denotes at edge node EN i Inner, task block request w jk Allocation to VMs l Carrying out treatment;
wherein r is ij Is an unmanned aerial vehicle terminal UE j To the edge node EN i Is measured in a physical distance of the mobile device, c is the propagation speed of the wireless channel;requesting w for a task block jk The calculated time delay at the edge node EN,is w jk At edge node EN i Virtual machine VM l Time delay of processing, w jk Indicating unmanned aerial vehicle terminal UE j The last kth task block request, q il (w jk ) Indicating w in a task block request sequence jk Sum of all task block requests, ω, previously pending jk Representing task Block requests w jk Load size of v il Representing edge nodes EN i Middle virtual machine VM l The processing rate of the CPU of (1); omega j'k' Load size, x, requested for a task block in a sequence of task block requests j'k' = i denotes that this task block request is allocated to the edge node EN i 。
5. The time delay and energy consumption-oriented power internet of things workload distribution method according to claim 4, wherein the method comprises the following steps:
step 2, the energy consumption of the unmanned aerial vehicle terminal UE is the sum of the energy consumption of all the unmanned aerial vehicle terminals UE, and the calculation formula of the energy consumption E of the unmanned aerial vehicle terminal UE is as follows:
wherein e is j For a single drone terminal UE j The energy consumption of (2) is reduced,for a single drone terminal UE j The energy consumption of the transmission of (2),for a single drone terminal UE j Energy consumption of movement of b jk Indicating the amount of data to be transmitted, e tc Representing a parameter of energy consumption of the transmitting circuit, e pa Representing a power amplification energy consumption parameter, wherein r is a transmission distance; e.g. of the type base Representing a basic kinetic energy consumption parameter, alpha j Is a basic motion energy consumption adjustment factor, independent of the workload distribution mode of the unmanned aerial vehicle terminal UE, e add Representing an additional kinetic energy consumption parameter, beta j Adjusting a factor for additional motion energy consumption;and distributing Euclidean distance between vectors for optimizing the working load of the front unmanned aerial vehicle terminal UE and the rear unmanned aerial vehicle terminal UE.
6. The time delay and energy consumption oriented power internet of things workload distribution method according to claim 5, wherein the method comprises the following steps:
the multi-objective optimization function of the power internet of things work load distribution established in the step 2 is as follows:
0≤ν il ≤ν i (16)
wherein D (x, v) represents the time delay of the unmanned aerial vehicle terminal UE; e (x) represents the energy consumption of the drone terminal UE;
x=(x jk ) JK×1 a task allocation matrix X = (X) for mapping a task block request in the drone terminal UE to the edge node EN jk ) J×K Is in the form of a vector, matrix element x jk Representing task Block requests w jk The number of the edge node EN assigned;
v=(ν il ) IL×1 and allocating a matrix V = (V) to the virtual machine VM in the edge node EN il ) I×L In the form of column vectors, matrix element v il Representing edge nodes EN i Middle virtual machine VM l The CPU processing rate of (1);
ν i representing the processing rate of a CPU in the edge node EN;
d j for a single drone terminal UE j The time delay of (1) is the sum of the network time delay and the calculation time delay;for unmanned terminal UE j A delay constraint threshold of (c);
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010079874.6A CN111314123B (en) | 2020-02-04 | 2020-02-04 | Time delay and energy consumption-oriented power Internet of things work load distribution method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010079874.6A CN111314123B (en) | 2020-02-04 | 2020-02-04 | Time delay and energy consumption-oriented power Internet of things work load distribution method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111314123A CN111314123A (en) | 2020-06-19 |
CN111314123B true CN111314123B (en) | 2022-11-04 |
Family
ID=71146953
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010079874.6A Active CN111314123B (en) | 2020-02-04 | 2020-02-04 | Time delay and energy consumption-oriented power Internet of things work load distribution method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111314123B (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111885651B (en) * | 2020-07-27 | 2022-06-07 | 中国联合网络通信集团有限公司 | Processing method and device for service concurrency of Internet of things |
CN111984408B (en) * | 2020-08-14 | 2021-04-20 | 昆山华泛信息服务有限公司 | Data cooperative processing method based on big data and edge computing and edge cloud platform |
CN112969157B (en) * | 2021-02-22 | 2022-10-18 | 重庆邮电大学 | Network load balancing method for unmanned aerial vehicle |
CN113630174B (en) * | 2021-08-31 | 2022-11-18 | 广东工业大学 | Unmanned aerial vehicle service data terminal transfer preprocessing method and scheduling method thereof |
CN114189874B (en) * | 2021-12-13 | 2024-04-16 | 国家电网有限公司 | 5G-based network construction method for power transmission line online monitoring system |
CN114567587B (en) * | 2022-03-14 | 2024-02-20 | 广州穗华能源科技有限公司 | Routing decision method for multi-user access edge computing terminal of power distribution Internet of things |
CN115001970A (en) * | 2022-05-26 | 2022-09-02 | 国网智能电网研究院有限公司 | Power service time delay optimization method and device |
CN116862152B (en) * | 2023-06-19 | 2024-03-05 | 广东电网有限责任公司汕尾供电局 | Load balancing-based task allocation method for power tower acceptance unmanned aerial vehicle |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103384272A (en) * | 2013-07-05 | 2013-11-06 | 华中科技大学 | Cloud service distributed data center system and load dispatching method thereof |
CN109067490A (en) * | 2018-09-29 | 2018-12-21 | 郑州航空工业管理学院 | Cellular Networks join lower multiple no-manned plane and cooperate with mobile edge calculations method for distributing system resource |
CN109814951A (en) * | 2019-01-22 | 2019-05-28 | 南京邮电大学 | The combined optimization method of task unloading and resource allocation in mobile edge calculations network |
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 |
-
2020
- 2020-02-04 CN CN202010079874.6A patent/CN111314123B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103384272A (en) * | 2013-07-05 | 2013-11-06 | 华中科技大学 | Cloud service distributed data center system and load dispatching method thereof |
CN109067490A (en) * | 2018-09-29 | 2018-12-21 | 郑州航空工业管理学院 | Cellular Networks join lower multiple no-manned plane and cooperate with mobile edge calculations method for distributing system resource |
CN109814951A (en) * | 2019-01-22 | 2019-05-28 | 南京邮电大学 | The combined optimization method of task unloading and resource allocation in mobile edge calculations network |
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 |
Non-Patent Citations (1)
Title |
---|
Workload Allocation Mechanism for Minimum Service Delay in Edge Computing-Based Power Internet of Things;Xudong Niu等;《IEEE Access》;20190531;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN111314123A (en) | 2020-06-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111314123B (en) | Time delay and energy consumption-oriented power Internet of things work load distribution method | |
CN110365753B (en) | Low-delay load distribution method and device for Internet of things service based on edge calculation | |
CN109885397B (en) | Delay optimization load task migration algorithm in edge computing environment | |
CN108182115B (en) | Virtual machine load balancing method in cloud environment | |
CN107172166B (en) | Cloud and mist computing system for industrial intelligent service | |
Mapetu et al. | A dynamic VM consolidation approach based on load balancing using Pearson correlation in cloud computing | |
Ren et al. | Coding-based large-scale task assignment for industrial edge intelligence | |
CN111132235B (en) | Mobile offload migration algorithm based on improved HRRN algorithm and multi-attribute decision | |
Wang et al. | Load balancing task scheduling based on multi-population genetic algorithm in cloud computing | |
CN111556516B (en) | Distributed wireless network task cooperative distribution method facing delay and energy efficiency sensitive service | |
CN107357652A (en) | A kind of cloud computing method for scheduling task based on segmentation sequence and standard deviation Dynamic gene | |
CN110795208A (en) | Mobile cloud computing self-adaptive virtual machine scheduling method based on improved particle swarm | |
CN112835684B (en) | Virtual machine deployment method for mobile edge computing | |
CN111131447A (en) | Load balancing method based on intermediate node task allocation | |
CN115967990A (en) | Classification and prediction-based border collaborative service unloading method | |
CN114356544A (en) | Parallel computing method and system facing edge cluster | |
CN114024970A (en) | Power internet of things work load distribution method based on edge calculation | |
CN115858675A (en) | Non-independent same-distribution data processing method based on federal learning framework | |
CN107454630B (en) | Load balancing method and load balancing router | |
CN111343602B (en) | Joint layout and task scheduling optimization method based on evolutionary algorithm | |
Zhang et al. | Dynamic DNN decomposition for lossless synergistic inference | |
CN114785692A (en) | Virtual power plant aggregation regulation and control communication network flow balancing method and device | |
CN114546609A (en) | DNN inference task batch scheduling method facing heterogeneous cluster | |
CN114650321A (en) | Task scheduling method for edge computing and edge computing terminal | |
CN111930435A (en) | Task unloading decision method based on PD-BPSO technology |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |