CN114519457A - Provincial intelligent energy service platform task scheduling method and system based on particle swarm - Google Patents

Provincial intelligent energy service platform task scheduling method and system based on particle swarm Download PDF

Info

Publication number
CN114519457A
CN114519457A CN202210100972.2A CN202210100972A CN114519457A CN 114519457 A CN114519457 A CN 114519457A CN 202210100972 A CN202210100972 A CN 202210100972A CN 114519457 A CN114519457 A CN 114519457A
Authority
CN
China
Prior art keywords
particle
particles
task scheduling
optimal
service platform
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
Application number
CN202210100972.2A
Other languages
Chinese (zh)
Inventor
张华鲁
石杰
张春平
张智成
李伟
张迎
邓博雅
马斌
王清明
周小飞
朱海
王德志
高志平
陈贵生
王喜鑫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nari Information and Communication Technology Co
Original Assignee
Nari Information and Communication Technology Co
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nari Information and Communication Technology Co filed Critical Nari Information and Communication Technology Co
Priority to CN202210100972.2A priority Critical patent/CN114519457A/en
Publication of CN114519457A publication Critical patent/CN114519457A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/08—Computing arrangements based on specific mathematical models using chaos models or non-linear system models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06313—Resource planning in a project environment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Strategic Management (AREA)
  • Biophysics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • General Health & Medical Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Biomedical Technology (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Development Economics (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • Game Theory and Decision Science (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Computational Mathematics (AREA)
  • Algebra (AREA)
  • Educational Administration (AREA)
  • Physiology (AREA)
  • Genetics & Genomics (AREA)
  • Mathematical Analysis (AREA)

Abstract

The invention discloses a provincial intelligent energy service platform task scheduling method and a provincial intelligent energy service platform task scheduling system based on particle swarm, belonging to the technical field of power scheduling, wherein the method comprises the following steps: performing particlization coding on a task scheduling sequence of the platform virtual machine to obtain a particle swarm corresponding to the task sequence in the task scheduling sequence, and performing parameter initialization; iteratively updating the speed and the position of the particle to search the optimal position of the particle, and stopping iteration until a preset convergence condition is met, wherein in the particle optimizing process, the speed and the position of the particle are updated according to the motion state of the particle and a preset strategy, and boundary processing is performed on the updated speed and position of the particle; determining the optimal position and the corresponding optimal fitness value of the particles obtained after iteration is stopped; and determining an optimal task scheduling sequence according to the optimal positions of the particles and the corresponding optimal fitness values. The invention improves the speed and the precision of scheduling the optimal tasks of the provincial intelligent energy service platform.

Description

Provincial intelligent energy service platform task scheduling method and system based on particle swarm
Technical Field
The invention belongs to the technical field of power scheduling, and particularly relates to a provincial intelligent energy service platform task scheduling method and system based on particle swarm.
Background
The cloud computing technology is a new technology which is popular in the internet industry at present, and is developed by combining traditional computers such as distributed computing, parallel computing, utility computing, network storage, virtualization, load balancing and the like with emerging network technologies. The cloud computing is introduced into the technical field of the smart power grid, and data resources and processor resources of the current system can be integrated under the condition that the hardware infrastructure of the existing power system is basically unchanged, so that the real-time control and advanced analysis capabilities of the power grid are greatly improved, and effective support is provided for the development of the smart power grid technology. The task scheduling problem of the provincial intelligent energy service platform is an NP complete problem, and in the prior art, the speed and the precision for obtaining the optimal task are low, so that the scheduling requirement is difficult to meet.
Disclosure of Invention
In order to overcome the defects in the prior art, the invention provides the provincial intelligent energy service platform task scheduling method and system based on the particle swarm, and the speed and the precision of optimal task scheduling of the provincial intelligent energy service platform are improved.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
in a first aspect, a task scheduling method for an intelligent energy service platform is provided, which includes: performing particlization coding on a task scheduling sequence of the platform virtual machine to obtain a particle swarm corresponding to the task sequence in the task scheduling sequence, and performing parameter initialization; iteratively updating the speed and the position of the particle to search the optimal position of the particle, and stopping iteration until a preset convergence condition is met, wherein in the particle optimizing process, the speed and the position of the particle are updated according to the motion state of the particle and a preset strategy, and boundary processing is performed on the updated speed and position of the particle; determining the optimal position and the corresponding optimal fitness value of the particles obtained after iteration is stopped; and determining an optimal task scheduling sequence according to the optimal positions of the particles and the corresponding optimal fitness values.
Further, the updating the speed and the position of the particle according to the motion state of the particle and a preset strategy includes: and if the particles are in an acceleration state, selecting a classical particle swarm optimization strategy to update the particle speed, if the particles are in a deceleration state, selecting a dimensional evolution strategy to update the particle speed, and updating the positions of the particles after the particle speed is updated.
Further, the classical particle swarm optimization strategy comprises:
Figure BDA0003492316100000021
where ω is the inertial weight, Vi tIs the velocity of the particle i at time t,
Figure BDA0003492316100000022
indicating the position of the ith particle at time t,
Figure BDA0003492316100000023
which represents the best position found by particle i up to time t, i.e. the local optimum,
Figure BDA0003492316100000024
the optimal positions of all particles at the moment t are shown, namely the global optimal positions; c. C1,c2Is a learning factor, r1,r2Is a random number between 0 and 1.
Further, the dimension evolution strategy comprises:
Figure BDA0003492316100000025
wherein the content of the first and second substances,
Figure BDA0003492316100000026
representing the velocity of the particle i in the d dimension,
Figure BDA0003492316100000027
representing the individual historical optimum of particle i in the d dimension,
Figure BDA0003492316100000028
representing the global historical optimum of particle i in the d dimension at time t.
Further, the method for determining the motion state of the particle includes: and calculating the acceleration of the particles at the current moment and the acceleration of the particles at the previous moment, comparing, and if the acceleration of the particles at the current moment is greater than the acceleration of the particles at the previous moment, indicating that the particles are in an acceleration state, otherwise, indicating that the particles are in a deceleration state.
Further, the boundary processing of the speed and the position of the particle includes: the velocity of the particles is limited using a boundary bounce strategy, which is expressed as:
Figure BDA0003492316100000031
wherein, Vi tDenotes the velocity, V, of the particle i at time tmaxAnd VminRepresenting the upper and lower limits of particle velocity. The position of the particle update is limited by adopting a boundary absorption strategy, which is expressed as follows:
Figure BDA0003492316100000032
wherein the content of the first and second substances,
Figure BDA0003492316100000033
denotes the position of the ith particle at time t, XmaxAnd XminRepresenting the upper and lower boundaries of the location, respectively.
Further, the preset convergence condition includes: setting the optimal fitness value of the particle swarm after n iterations as
Figure BDA0003492316100000034
Setting population m times before<n) has an optimum fitness value of
Figure BDA0003492316100000035
Given a convergence threshold u, if
Figure BDA0003492316100000036
The algorithm is determined to have converged and otherwise not converged.
Further, if the algorithm has converged, the algorithm will,judging whether the particles are premature or not, wherein the method for judging whether the particles are premature or not comprises the following steps: let fiIs the fitness value of the ith particle,
Figure BDA0003492316100000037
is the current average fitness of the population, and the variance of the fitness of the population is defined as sigma2The formula is as follows:
Figure BDA0003492316100000038
wherein f is called a normalization factor, and the value of f is determined according to the following formula:
Figure BDA0003492316100000039
setting a threshold value C of the variance of the fitness when sigma is2<C, it is described that the particles are in an aggregated state at this time, and it is necessary to perform precocity processing to determine the algorithm state as precocity. Otherwise, the algorithm is not in the early maturing state.
Further, if the particles are in the premature state, chaotic variables are generated based on Logitics chaotic mapping to update the positions of the particles, so that the particles are helped to get rid of the premature state.
In a second aspect, a task scheduling system for a smart energy service platform is provided, which includes a processor and a storage device, where multiple instructions are stored in the storage device, and the processor is configured to load and execute the steps of the method according to the first aspect.
Compared with the prior art, the invention has the following beneficial effects:
(1) the invention carries out particlization coding on the task scheduling sequence of the platform virtual machine, and selects different strategies to update the speed and the position of the particles according to the motion state of the particles; carrying out boundary processing on the speed and the position of the particles; the position of the optimal particle and the corresponding optimal fitness value are output after convergence, so that an optimal task scheduling sequence is obtained, and the speed and the precision of optimal task scheduling of the provincial intelligent energy service platform are improved;
(2) the invention provides a method for judging the motion state of particles and combines a dimension evolution strategy to specifically update the dimension of each particle one by one, so as to perfect the traditional scheme, thereby effectively improving the searching capability of the algorithm, avoiding the algorithm from falling into the local optimal solution by adding an early judgment and chaotic disturbance mechanism, solving the problem of easy falling into the local optimal solution and achieving the effect of improving the accuracy of the algorithm.
Drawings
Fig. 1 is a schematic main flowchart of a provincial intelligent energy service platform task scheduling method based on particle swarm provided in an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
The first embodiment is as follows:
a heuristic adaptive particle swarm-based provincial smart energy service platform task scheduling method comprises the following steps: performing particlization coding on a task scheduling sequence of the platform virtual machine according to the scale of the provincial intelligent energy service platform task to obtain a particle swarm corresponding to the task sequence in the task scheduling sequence, and performing parameter initialization; iteratively updating the speed and the position of the particle to search the optimal position of the particle, and stopping iteration until a preset convergence condition is met, wherein in the particle optimizing process, the speed and the position of the particle are updated according to the motion state of the particle and a preset strategy, and boundary processing is performed on the updated speed and position of the particle; determining the optimal position and the corresponding optimal fitness value of the particles obtained after iteration is stopped; and determining an optimal task scheduling sequence according to the optimal positions of the particles and the corresponding optimal fitness values.
As shown in fig. 1, the task scheduling method for the provincial intelligent energy service platform based on the heuristic adaptive particle swarm specifically adopts the following technical scheme.
The method comprises the following steps: and performing particlization coding on the task scheduling sequence of the platform virtual machine according to the scale of the provincial intelligent energy service platform task to obtain a particle swarm corresponding to the task sequence in the task scheduling sequence, and performing parameter initialization.
Let the number of tasks of the provincial intelligent energy service platform be D, and the task scheduling sequence is represented as a particle, that is, the particle has a D-dimensional solution space. Let X be the initial number of particles Ni＝(xi1,xi1,...,xiD),i∈[1,N]Denotes the position of the ith particle, Xbest,i＝Xb,i＝(xb1,i,xb1,i,...,xbD,i) Representing the optimal position searched by a single particle i. In the same way, Xbest＝Xg＝(xg1,xg1,...,xgD) The optimal positions searched by all the particles are shown, namely the optimal task scheduling scheme calculated currently. And the particles continuously update the two optimal positions through continuous iteration so as to continuously approach the optimal task scheduling scheme. Specifically, assuming that ten tasks are to be optimized for scheduling, the dimension D of the particle is set to 10, and the value of the particle in each dimension represents the sequence number of the virtual machine node assigned by the task:
TABLE 1 provincial-level intelligent energy service platform task code
Figure BDA0003492316100000051
Figure BDA0003492316100000061
As shown in table 1, the task code of the provincial intelligent energy service platform is shown in the embodiment.
Step two: and iteratively updating the speed and the position of the particle to search the optimal position of the particle, and stopping iteration until a preset convergence condition is met, wherein in the particle optimizing process, the speed and the position of the particle are updated according to the motion state of the particle and a preset strategy.
The method for judging the motion state of the particles comprises the following steps: and calculating the acceleration of the particles at the current moment and the acceleration of the particles at the previous moment, comparing, and if the acceleration of the particles at the current moment is greater than the acceleration of the particles at the previous moment, indicating that the particles are in an acceleration state, otherwise, indicating that the particles are in a deceleration state.
If the particles are in an accelerated state, a classical particle swarm optimization strategy is selected to update the particle speed, and the method comprises the following steps:
Figure BDA0003492316100000062
where ω is the inertial weight, Vi tIs the velocity of the particle i at time t,
Figure BDA0003492316100000063
indicating the position of the ith particle at time t,
Figure BDA0003492316100000064
which represents the best position found by particle i up to time t, i.e. the local optimum,
Figure BDA0003492316100000065
the optimal positions of all particles at the moment t are shown, namely the global optimal positions; c. C1,c2Is a learning factor, r1,r2Is a random number between 0 and 1.
If the particle is in a deceleration state, selecting a dimension evolution strategy to update the velocity of the particle (the initial particle is in an acceleration state), and after the velocity of the particle is updated, updating the position of the particle, wherein the method comprises the following steps:
Figure BDA0003492316100000066
wherein the content of the first and second substances,
Figure BDA0003492316100000067
representing the velocity of particle i in the d dimension at time t,
Figure BDA0003492316100000068
representing the individual historical optimum of particle i in the d dimension,
Figure BDA0003492316100000069
representing the global historical optimum of particle i in the d dimension at time t.
After the particle velocity is updated, the position of the particle at the time t +1 can be updated as follows:
Figure BDA00034923161000000610
wherein the content of the first and second substances,
Figure BDA0003492316100000071
denotes the position of the ith particle at time t, Vi t+1Representing the velocity of the particle at time t +1, then
Figure BDA0003492316100000072
Denoted as the position of the ith particle at time t + 1.
Step three: in order to prevent the particles from exceeding the value range of the knowledge space in the searching process, boundary processing is carried out on the updated speed and position of the particles.
Specifically, the velocity of the particles is limited using a boundary bounce strategy, which is expressed as:
Figure BDA0003492316100000073
wherein, Vi tDenotes the velocity, V, of the particle i at time tmaxAnd VminRepresenting the upper and lower limits of particle velocity.
The position of the particle update is limited by using a boundary absorption strategy, which is expressed as:
Figure BDA0003492316100000074
wherein the content of the first and second substances,
Figure BDA0003492316100000075
denotes the position of the ith particle at time t, XmaxAnd XminRepresenting the upper and lower boundaries of the location, respectively.
And determining the optimal position and the corresponding optimal fitness value of the particle obtained after the iteration is stopped, and determining the optimal task scheduling sequence according to the optimal position and the corresponding optimal fitness value of the particle.
Step four: and judging whether convergence is achieved or not, if not, entering the step five, and if not, entering the step six.
Whether the algorithm is converged or not is judged based on the fitness, the fitness of the algorithm is reflected by the position of each individual, and the relative position of the particles is judged through the global change of the values of all the particles, so that whether the algorithm is converged or not is judged.
The method for judging whether convergence occurs comprises the following steps: setting the optimal fitness value of the particle swarm after n iterations as
Figure BDA0003492316100000076
Setting population m times before<n) has an optimum fitness value of
Figure BDA0003492316100000077
Given a convergence threshold u, if
Figure BDA0003492316100000078
The algorithm is determined to have converged and otherwise not converged.
Step five: judging the speed state of the particles, returning to the step two, specifically, comparing the acceleration of the particles at the current moment with the historical acceleration at the previous moment, and if V is the speed state of the particles, judging whether the acceleration of the particles at the current moment is the historical acceleration at the previous momenti t+1-Vi t3Vi t-Vi t-1Then it means that the current particle is in the acceleration state, if Vi t+1-Vi t<Vi t-Vi t-1The surface is currently in a deceleration state.
Step six: and (4) judging whether the particles are premature, if so, entering a seventh step, and otherwise, entering an eighth step.
The method for judging whether the particles are premature comprises the following steps: let f beiIs the fitness value of the ith particle,
Figure BDA0003492316100000081
is the current average fitness of the population, and the variance of the fitness of the population is defined as sigma2The formula is as follows:
Figure BDA0003492316100000082
wherein f is called a normalization factor, and the value of f is determined according to the following formula:
Figure BDA0003492316100000083
setting a threshold value C of the variance of the fitness when sigma is2<C, it is described that the particles are in an aggregated state at this time, and it is necessary to perform precocity processing to determine the algorithm state as precocity. Otherwise, the algorithm is not in the early maturing state.
Step seven: and performing chaotic disturbance treatment, setting the speed of the particles as an acceleration state, and entering the step two.
Specifically, chaotic perturbations help a particle escape from the current location when the particle falls into a locally optimal solution. The logistic chaos mapping idea is adopted to generate a chaos variable, and a dynamic model of the chaos variable is defined as follows:
yk+1＝σ×yk(1+yk) (8)
wherein, sigma is a control parameter and sigma belongs to [0,4 ]]，yk∈[0,1]. When 3.5699<When the sigma is less than or equal to 4, the system is in a chaotic state. In addition, when σ is 4, the system is in a completely chaotic state. Then, introducing the chaos thought into a particle swarm optimization algorithm by using a carrier mapping method to obtain a chaos disturbance parameter of the particle:
li＝4l(1+l) (9)
wherein liIs the chaotic disturbance parameter of the particle i, l is a random number between 0 and 1, and l belongs to [0,1 ]]. Randomly selecting m particles trapped into stagnation from the N particles to carry out chaotic disturbance, and obtaining m new particles, wherein the expression is as follows:
Figure BDA0003492316100000091
wherein, XiAt the position of the particle i, Xi,newFor the perturbed acquisition of particle i in a new position, XmaxAnd XminRespectively representing the upper and lower boundaries of the location, liAnd the chaotic disturbance parameter is corresponding to the particle i. And 8: and outputting the optimal particle position and the corresponding optimal fitness value and finishing.
All dimensions of each particle of the traditional particle swarm optimization algorithm are considered as a whole to be updated, so that the fitness value represented by the particle is continuously close to the optimal solution, but the value of each dimension is not towards the optimal direction at a uniform speed. Conventional strategies, while fast convergence is possible, tend to result in a locally optimal solution being trapped. The research improves the traditional scheme by providing a judgment method of the motion state of the particles and combining a dimension evolution strategy to specifically update the dimension of each particle one by one, thereby effectively improving the searching capability of the algorithm. And the algorithm is prevented from falling into the local optimal solution by adding the premature judgment and the chaotic disturbance mechanism, so that the problem of easy falling into the local optimal solution is solved, and the effect of improving the accuracy of the algorithm is achieved. Therefore, the speed and the precision of scheduling the optimal tasks of the provincial intelligent energy service platform are improved.
The second embodiment:
the embodiment provides a task scheduling system of a provincial smart energy service platform based on heuristic adaptive particle swarm, which comprises a processor and a storage device, wherein the storage device stores a plurality of instructions for the processor to load and execute the steps of the method of the first embodiment.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.

Claims (10)

1. A task scheduling method of an intelligent energy service platform is characterized by comprising the following steps:
performing particlization coding on a task scheduling sequence of the platform virtual machine to obtain a particle swarm corresponding to the task sequence in the task scheduling sequence, and performing parameter initialization;
iteratively updating the speed and the position of the particle to search the optimal position of the particle, and stopping iteration until a preset convergence condition is met, wherein in the particle optimizing process, the speed and the position of the particle are updated according to the motion state of the particle and a preset strategy, and boundary processing is performed on the updated speed and position of the particle;
determining the optimal position and the corresponding optimal fitness value of the particles obtained after iteration is stopped;
and determining an optimal task scheduling sequence according to the optimal positions of the particles and the corresponding optimal fitness values.
2. The intelligent energy service platform task scheduling method of claim 1, wherein the updating the speed and the position of the particles according to the motion state of the particles and a preset strategy comprises: and if the particles are in an acceleration state, selecting a classical particle swarm optimization strategy to update the speed of the particles, if the particles are in a deceleration state, selecting a dimension evolution strategy to update the speed of the particles, and updating the positions of the particles after the speed of the particles is updated.
3. The intelligent energy service platform task scheduling method according to claim 2, wherein the classical particle swarm optimization strategy comprises:
Figure FDA0003492316090000011
where ω is the inertial weight, Vi tIs the velocity of the particle i at time t,
Figure FDA0003492316090000012
indicates the position of the ith particle,
Figure FDA0003492316090000013
indicating the best position found for particle i, i.e. the local optimum,
Figure FDA0003492316090000014
representing the optimal position in all particles, i.e. global optimal; c. C1,c2Is a learning factor, r1,r2Is a random number between 0 and 1.
4. The intelligent energy service platform task scheduling method of claim 2, wherein the dimension evolution strategy comprises:
Figure FDA0003492316090000021
wherein, the first and the second end of the pipe are connected with each other,
Figure FDA0003492316090000022
representing the velocity of the particle i in the d dimension,
Figure FDA0003492316090000023
representing the individual historical optimum of particle i in the d dimension,
Figure FDA0003492316090000024
representing the global historical optimum of particle i in the d dimension.
5. The intelligent energy service platform task scheduling method of claim 1, wherein the method for determining the motion state of the particles comprises: and calculating the acceleration of the particles at the current moment and the acceleration of the particles at the previous moment, comparing, and if the acceleration of the particles at the current moment is greater than the acceleration of the particles at the previous moment, indicating that the particles are in an acceleration state, otherwise, indicating that the particles are in a deceleration state.
6. The intelligent energy service platform task scheduling method of claim 1, wherein the boundary processing of the velocity and the position of the particles comprises:
the velocity of the particles is limited using a boundary bounce strategy, which is expressed as:
Figure FDA0003492316090000025
wherein, Vi tDenotes the velocity, V, of the particle i at time tmaxAnd VminRepresenting the upper and lower limits of particle velocity. The position of the particle update is limited by using a boundary absorption strategy, which is expressed as:
Figure FDA0003492316090000026
wherein the content of the first and second substances,
Figure FDA0003492316090000027
denotes the position of the ith particle at time t, XmaxAnd XminRepresenting the upper and lower boundaries of the location, respectively.
7. The intelligent energy service platform task scheduling method of claim 1, wherein the predetermined convergence condition comprises: setting the optimal fitness value of the particle swarm after n iterations as
Figure FDA0003492316090000028
Setting population m times before<n) has an optimum fitness value of
Figure FDA0003492316090000029
Given a convergence threshold u, if
Figure FDA00034923160900000210
The algorithm is determined to have converged and otherwise not converged.
8. The intelligent energy service platform task scheduling method of claim 7, wherein if the algorithm has converged, determining whether the particles are premature, the method of determining whether the particles are premature comprises: let fiIs the fitness value of the ith particle,
Figure FDA0003492316090000031
is the current average fitness of the population, and the variance of the fitness of the population is defined as sigma2The formula is as follows:
Figure FDA0003492316090000032
wherein f is called a normalization factor, and the value of f is determined according to the following formula:
Figure FDA0003492316090000033
setting a threshold value C of the variance of the fitness when sigma is2If < C, the particles are in an aggregation state at this time, premature processing is required, and the algorithm state is determined to be premature. Otherwise, the algorithm is not in the early maturing state.
9. The intelligent energy service platform task scheduling method of claim 8, wherein if the particles are in an early maturing state, generating chaotic variables based on Logitics chaotic mapping to update positions of the particles to help the particles get rid of the early maturing state.
10. A task scheduling system of a smart energy service platform, comprising a processor and a storage device, wherein the storage device stores a plurality of instructions for the processor to load and execute the steps of the method according to any one of claims 1 to 9.
CN202210100972.2A 2022-01-27 2022-01-27 Provincial intelligent energy service platform task scheduling method and system based on particle swarm Pending CN114519457A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210100972.2A CN114519457A (en) 2022-01-27 2022-01-27 Provincial intelligent energy service platform task scheduling method and system based on particle swarm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210100972.2A CN114519457A (en) 2022-01-27 2022-01-27 Provincial intelligent energy service platform task scheduling method and system based on particle swarm

Publications (1)

Publication Number Publication Date
CN114519457A true CN114519457A (en) 2022-05-20

Family

ID=81596197

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210100972.2A Pending CN114519457A (en) 2022-01-27 2022-01-27 Provincial intelligent energy service platform task scheduling method and system based on particle swarm

Country Status (1)

Country Link
CN (1) CN114519457A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116880163A (en) * 2023-09-07 2023-10-13 北京英沣特能源技术有限公司 Intelligent data center cold source regulation and control method and system

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116880163A (en) * 2023-09-07 2023-10-13 北京英沣特能源技术有限公司 Intelligent data center cold source regulation and control method and system
CN116880163B (en) * 2023-09-07 2023-12-05 北京英沣特能源技术有限公司 Intelligent data center cold source regulation and control method and system

Similar Documents

Publication Publication Date Title
CN110851272B (en) Cloud task scheduling method based on phagocytic particle swarm genetic hybrid algorithm
CN108182115B (en) Virtual machine load balancing method in cloud environment
CN108694077B (en) Distributed system task scheduling method based on improved binary system bat algorithm
CN110418353B (en) Edge computing server placement method based on particle swarm algorithm
CN107506865B (en) Load prediction method and system based on LSSVM optimization
CN107317699B (en) Dynamic ant colony rapid optimization method of cloud manufacturing service combination
CN111915079B (en) Hybrid KNN wind power prediction method and system
CN112132469B (en) Reservoir group scheduling method and system based on multiple group cooperation particle swarm algorithm
Li et al. Chaotic particle swarm optimization algorithm based on adaptive inertia weight
CN114519457A (en) Provincial intelligent energy service platform task scheduling method and system based on particle swarm
CN111681258A (en) Hybrid enhanced intelligent trajectory prediction method and device based on hybrid wolf optimization SVM
CN113316116A (en) Vehicle calculation task unloading method based on multi-arm gambling machine
CN113391894A (en) Optimization method of optimal hyper-task network based on RBP neural network
CN112257897B (en) Electric vehicle charging optimization method and system based on improved multi-target particle swarm
CN109193662B (en) Most dangerous load margin calculation method and system considering unbalanced power sharing
Ding et al. A task scheduling algorithm for heterogeneous systems using aco
CN116108948A (en) Wind-solar and V2G charging station site selection and volume fixation optimization method for power distribution network
CN110334797A (en) A kind of calculation method and system based on parallel enhancing search particle swarm optimization algorithm
CN106611280A (en) Imperialism competition algorithm based on real variable function side distance
CN116431281A (en) Virtual machine migration method based on whale optimization algorithm
Tan et al. A fast and stable forecasting model to forecast power load
CN116089083A (en) Multi-target data center resource scheduling method
CN115271254A (en) Short-term wind power prediction method for optimizing extreme learning machine based on gull algorithm
CN115496133A (en) Density data stream clustering method based on self-adaptive online learning
CN114531665A (en) Wireless sensor network node clustering method and system based on Laiwei flight

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