CN114997659B - Resource scheduling model construction method and system based on dynamic multi-objective optimization - Google Patents

Resource scheduling model construction method and system based on dynamic multi-objective optimization Download PDF

Info

Publication number
CN114997659B
CN114997659B CN202210645045.9A CN202210645045A CN114997659B CN 114997659 B CN114997659 B CN 114997659B CN 202210645045 A CN202210645045 A CN 202210645045A CN 114997659 B CN114997659 B CN 114997659B
Authority
CN
China
Prior art keywords
optimization
resource scheduling
scheduling model
time
task
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
Application number
CN202210645045.9A
Other languages
Chinese (zh)
Other versions
CN114997659A (en
Inventor
李昕
张靖凯
张君
周明宇
黄江帆
汤旭晶
冯龙祥
冯玉龙
李骁
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.)
Zhejiang Electronic Information Product Inspection And Research Institute
Wuhan University of Technology WUT
Original Assignee
Zhejiang Electronic Information Product Inspection And Research Institute
Wuhan University of Technology WUT
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 Zhejiang Electronic Information Product Inspection And Research Institute, Wuhan University of Technology WUT filed Critical Zhejiang Electronic Information Product Inspection And Research Institute
Priority to CN202210645045.9A priority Critical patent/CN114997659B/en
Publication of CN114997659A publication Critical patent/CN114997659A/en
Application granted granted Critical
Publication of CN114997659B publication Critical patent/CN114997659B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06Q10/06—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
    • G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
    • 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
    • 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Game Theory and Decision Science (AREA)
  • General Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides a resource scheduling model construction method and a system based on dynamic multi-objective optimization, wherein the method comprises the following steps: constructing a resource scheduling model based on the task optimization total period, the task optimization subcycle, the optimization target model, the decision variable, the time-invariant constraint condition and the time-variant constraint condition; determining variable parameters affecting a resource scheduling model based on historical data, dividing the variable parameters, and generating a variable parameter set; performing off-line multi-objective optimization on the resource scheduling model based on an evolutionary multi-objective optimization algorithm and a plurality of variable parameter subsets, and determining the pareto optimal scheme set of each task optimization sub-period and the sensitivity of each resource optimization target model to decision variables; and storing the pareto optimal scheme set as a reference vector into an offline knowledge base, obtaining a simplified resource scheduling model according to the sensitivity, and storing the simplified resource scheduling model into the offline knowledge base. The invention improves the efficiency of multi-objective optimization of the resource scheduling task.

Description

Resource scheduling model construction method and system based on dynamic multi-objective optimization
Technical Field
The invention relates to the technical field of dynamic multi-objective optimization, in particular to a resource scheduling model construction method and system based on dynamic multi-objective optimization.
Background
The periodical multi-target resource scheduling task is commonly used in various occasions, such as network resource allocation, ship energy management, micro-grid optimized operation and the like, and is mainly characterized in that: more than one optimization target is needed, but the real-time requirement is high and the computing resources are often limited; the external parameters affecting the optimization effect have time variability, and show a certain similarity in the same optimization sub-period. Therefore, the corresponding optimization model has the characteristics of more objective functions, contradiction among the objective functions, unfixed constraint conditions and the like, so that the search space is complex, the feasible domain dynamically changes along with time, and the complete Pareto (Pareto) front edge is difficult to obtain.
In recent years, with the development of computer technology, advantages of heuristic optimization algorithms (such as evolutionary multi-objective optimization algorithms) in solving complex multi-objective optimization problems are gradually highlighted. However, for periodic multi-objective resource scheduling tasks, the efficiency of multi-objective optimization of resource scheduling tasks by evolving a multi-objective optimization algorithm is low due to limited online computing resources.
Disclosure of Invention
In view of the foregoing, it is necessary to provide a method and a system for constructing a resource scheduling model based on dynamic multi-objective optimization, so as to solve the technical problem in the prior art that the efficiency of performing multi-objective optimization on a resource scheduling task by evolving a multi-objective optimization algorithm is low due to effective on-line computing resources.
In one aspect, the invention provides a resource scheduling model construction method based on dynamic multi-objective optimization, which comprises the following steps:
determining a task optimization total period, a plurality of task optimization subcycles, a plurality of resource optimization target models, decision variables, time-invariant constraint conditions and time-variant constraint conditions of a resource scheduling task, and constructing a resource scheduling model based on the task optimization total period, the plurality of task optimization subcycles, the plurality of optimization target models, the decision variables, the time-invariant constraint conditions and the time-variant constraint conditions;
Determining variable parameters affecting the resource scheduling model based on historical data, dividing the variable parameters based on the task optimization total period and the plurality of task optimization subcycles, and generating a variable parameter set, wherein the variable parameter set comprises a plurality of variable parameter subsets which are in one-to-one correspondence with the plurality of task optimization subcycles;
Performing off-line multi-objective optimization on the resource scheduling model based on an evolutionary multi-objective optimization algorithm and the variable parameter subsets, and determining the pareto optimal scheme set of each task optimization sub-period and the sensitivity of each resource optimization target model in the resource optimization target models to the decision variables;
and storing the pareto optimal scheme set as a reference vector to an offline knowledge base, simplifying the resource scheduling model according to the sensitivity to obtain a simplified resource scheduling model, and storing the simplified resource scheduling model to the offline knowledge base.
In some possible implementations, the time-invariant constraints include time-invariant equality constraints and time-invariant inequality constraints, the time-variant constraints including time-variant equality constraints and time-variant inequality constraints; the resource scheduling model is as follows:
minF(x,T)＝{fa(x),a＝1,2,...,N}
x＝(xm,m＝1,2,...,M)∈X
Wherein, minF (x, T) is a resource scheduling model; f a (x) is a optimization target models; n is the total number of the plurality of optimization target models; x is a decision vector; the T task is optimizing the total period duration; t is the time length of the task optimization subcycle; x m is the mth decision variable; m is the total number of decision variables; Is the i 1 th time-invariant inequality constraint condition; /(I) A constraint for the j 1 th time-invariant equation; /(I)Is the i 2 th time-varying inequality constraint; /(I)A 2 th time-varying equality constraint; p 1 is the total number of time-invariant inequality constraints; q 1 is the total number of time-invariant equation constraints; p 2 is the total number of time-varying inequality constraints; q 2 is the total number of time-varying equation constraints; x is the feasible region.
In some possible implementations, the variable parameter set is:
H(t)＝{H(t1),H(t2),...,H(tk),...,H(tK)}
H(tk)＝{H1(tk),H2(tk),...,Hr(tk),...,HR(tk)}
Hr(tk)＝{hr(tk,1),hr(tk,2),...,hr(tk,l),...,hr(tk,L)}
Wherein H (t) is a variable parameter set; h (t k) is a subset of variable parameters within the task optimization subcycle t k; k is the number of variable parameter subsets; h r(tk) optimizing a set of the r-th variable parameters in the sub-period for the task; h r(tk,l) is the parameter value of the r-th variable parameter in the l-th task optimization sub-period t k; l is the number of sub-periods for the same task optimization.
In some possible implementations, the sensitivity is:
Wherein Δo is sensitivity; o max is the maximum value of the resource optimization target model in the offline multi-target optimization process of the resource scheduling model, and O min is the minimum value of the resource optimization target model in the offline multi-target optimization process of the resource scheduling model.
In some possible implementations, the simplifying the resource scheduling model according to the sensitivity to obtain a simplified resource scheduling model includes:
Judging whether the sensitivity is smaller than a threshold sensitivity or not;
If the sensitivity is smaller than the threshold sensitivity, the resource optimization target model corresponding to the sensitivity is a resource optimization target model to be deleted, and the resource optimization target model to be deleted is deleted to obtain the simplified resource scheduling model;
And if the sensitivity is greater than or equal to the threshold sensitivity, the resource scheduling model is the simplified resource scheduling model.
In some possible implementations, the method for constructing a resource scheduling model based on dynamic multi-objective optimization further includes:
Acquiring a real-time optimization sub-period, and calling the simplified resource scheduling model and the reference vector from the offline knowledge base based on the real-time optimization sub-period;
determining a target reference vector from the reference vectors based on a vector determination model;
And carrying out real-time multi-objective optimization on the simplified resource scheduling model based on the evolutionary multi-objective optimization algorithm and the objective reference vector, determining a real-time pareto optimal solution set, and carrying out resource scheduling based on the real-time pareto optimal solution set.
In some possible implementations, the determining a target reference vector from the reference vectors based on the vector determination model includes:
determining a number of target reference vectors of a model from the reference vectors based on the vectors;
The target reference vector is randomly determined from the reference vectors based on the number of target reference vectors.
In some possible implementations, the method for constructing a resource scheduling model based on dynamic multi-objective optimization further includes:
And storing the real-time pareto optimal solution set as a real-time reference vector to the offline knowledge base.
In some possible implementations, the vector determination model is:
Wherein S ref is the number of target reference vectors; g max is the maximum number of iterations; g current is the current iteration number; To round operators.
On the other hand, the invention also provides a resource scheduling model construction system based on dynamic multi-objective optimization, which comprises the following steps:
The resource scheduling model construction module is used for determining a task optimization total period, a plurality of task optimization subcycles, a plurality of resource optimization target models, decision variables, time-invariant constraint conditions and time-variant constraint conditions of a resource scheduling task and constructing a resource scheduling model based on the task optimization total period, the plurality of task optimization subcycles, the plurality of optimization target models, the decision variables, the time-invariant constraint conditions and the time-variant constraint conditions;
A variable parameter set determining module, configured to determine variable parameters affecting the resource scheduling model based on historical data, and divide the variable parameters based on the task optimization total period and the plurality of task optimization sub-periods, to generate a variable parameter set, where the variable parameter set includes a plurality of variable parameter subsets corresponding to the plurality of task optimization sub-periods one to one;
The multi-objective optimization module is used for carrying out off-line multi-objective optimization on the resource scheduling model based on an evolutionary multi-objective optimization algorithm and the variable parameter subsets, and determining the pareto optimal scheme set of each task optimization sub-period and the sensitivity of each resource optimization target model in the resource optimization target models to the decision variables;
And the simplified resource scheduling model construction module is used for storing the pareto optimal scheme set as a reference vector to an offline knowledge base, simplifying the resource scheduling model according to the sensitivity to obtain a simplified resource scheduling model, and storing the simplified resource scheduling model to the offline knowledge base.
The beneficial effects of adopting the embodiment are as follows: according to the resource scheduling model construction method based on dynamic multi-objective optimization, the resource scheduling model is constructed, the plurality of variable parameter subsets are determined based on historical data, then the off-line multi-objective optimization is carried out on the resource scheduling model based on the evolutionary multi-objective optimization algorithm and the plurality of variable parameter subsets, so that the pareto optimal scheme set is obtained, on-line computing resources can be saved, when on-line real-time optimization is needed for the resource scheduling task, the resource scheduling model and the reference vector can be directly called from the off-line knowledge base, and the efficiency of multi-objective optimization on the resource scheduling task is improved.
Further, the invention simplifies the resource scheduling model according to the sensitivity to obtain the simplified resource scheduling model, thereby further reducing the complexity of the resource scheduling model, saving the online computing resources and further improving the efficiency of multi-objective optimization of the resource scheduling task.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow diagram of an embodiment of a method for constructing a resource scheduling model based on dynamic multi-objective optimization provided by the present invention;
FIG. 2 is a schematic diagram of the structure of an embodiment of a variable parameter subset according to the present invention;
FIG. 3 is a flow chart of the embodiment of S104 in FIG. 1 according to the present invention;
FIG. 4 is a flow chart of an embodiment of performing multi-objective optimization on a resource scheduling task based on a simplified resource scheduling model according to the present invention;
FIG. 5 is a flowchart illustrating the process of S402 of FIG. 4 according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of an embodiment of a system for constructing a resource scheduling model based on dynamic multi-objective optimization according to the present invention;
fig. 7 is a schematic structural diagram of an embodiment of an electronic device according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the schematic drawings are not drawn to scale. A flowchart, as used in this disclosure, illustrates operations implemented according to some embodiments of the present invention. It should be appreciated that the operations of the flow diagrams may be implemented out of order and that steps without logical context may be performed in reverse order or concurrently. Moreover, one or more other operations may be added to or removed from the flow diagrams by those skilled in the art under the direction of the present disclosure.
Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor systems and/or microcontroller systems.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
The embodiment of the invention provides a resource scheduling model construction method and system based on dynamic multi-objective optimization, which are respectively described below.
FIG. 1 is a schematic flow chart of one embodiment of a method for constructing a resource scheduling model based on dynamic multi-objective optimization, as shown in FIG. 1, wherein the method for constructing the resource scheduling model based on dynamic multi-objective optimization comprises the following steps:
S101, determining a task optimization total period, a plurality of task optimization subcycles, a plurality of resource optimization target models, decision variables, time-invariant constraint conditions and time-variant constraint conditions of a resource scheduling task, and constructing a resource scheduling model based on the task optimization total period, the plurality of task optimization subcycles, the plurality of optimization target models, the decision variables, the time-invariant constraint conditions and the time-variant constraint conditions;
S102, determining variable parameters affecting a resource scheduling model based on historical data, dividing the variable parameters based on a task optimization total period and a plurality of task optimization sub-periods, and generating a variable parameter set, wherein the variable parameter set comprises a plurality of variable parameter subsets which are in one-to-one correspondence with the plurality of task optimization sub-periods;
S103, performing off-line multi-objective optimization on the resource scheduling model based on an evolutionary multi-objective optimization algorithm and a plurality of variable parameter subsets, and determining the pareto optimal scheme set of each task optimization sub-period and the sensitivity of each resource optimization target model in a plurality of resource optimization target models to decision variables;
S104, storing the pareto optimal scheme set as a reference vector to an offline knowledge base, simplifying the resource scheduling model according to sensitivity to obtain a simplified resource scheduling model, and storing the simplified resource scheduling model to the offline knowledge base.
Compared with the prior art, the resource scheduling model construction method based on dynamic multi-objective optimization provided by the embodiment of the invention has the advantages that the resource scheduling model is constructed, a plurality of variable parameter subsets are determined based on historical data, then the resource scheduling model is subjected to off-line multi-objective optimization based on an evolutionary multi-objective optimization algorithm and the plurality of variable parameter subsets, so that the pareto optimal scheme set is obtained, on-line computing resources can be saved, when on-line real-time optimization of the resource scheduling task is required, the resource scheduling model and the reference vector can be directly called from an off-line knowledge base, and the efficiency of multi-objective optimization of the resource scheduling task is improved.
Further, the embodiment of the invention simplifies the resource scheduling model according to the sensitivity to obtain the simplified resource scheduling model, thereby further reducing the complexity of the resource scheduling model, saving the online computing resources and further improving the efficiency of multi-objective optimization of the resource scheduling task.
It should be understood that: the resource scheduling model includes, but is not limited to, a micro-grid resource scheduling model, a network resource scheduling model, and a ship energy resource scheduling model.
In some embodiments of the invention, the time-invariant constraint comprises a time-invariant equality constraint and a time-invariant inequality constraint, and the time-variant constraint comprises a time-variant equality constraint and a time-variant inequality constraint; the resource scheduling model is as follows:
minF(x,T)＝{fa(x),a＝1,2,...,N}
x＝(xm,m＝1,2,...,M)∈X
Wherein, minF (x, T) is a resource scheduling model; f a (x) is a optimization target models; n is the total number of the plurality of optimization target models; x is a decision vector; the T task is optimizing the total period duration; t is the time length of the task optimization subcycle; x m is the mth decision variable; m is the total number of decision variables; Is the i 1 th time-invariant inequality constraint condition; /(I) A constraint for the j 1 th time-invariant equation; /(I)Is the i 2 th time-varying inequality constraint; /(I)A 2 th time-varying equality constraint; p 1 is the total number of time-invariant inequality constraints; q 1 is the total number of time-invariant equation constraints; p 2 is the total number of time-varying inequality constraints; q 2 is the total number of time-varying equation constraints; x is the feasible region.
In a specific embodiment of the present invention, when the resource scheduling model is applied to the micro-grid, the resource scheduling model includes three optimization target models, namely an economic index optimization model, an environmental protection index optimization model and a safety index optimization model, and the decision vector is composed of the power output values of all the power supplies in the micro-grid, as shown in the following formula:
x＝(x1,x2,x3,x4,x5,x6)
Wherein x 1,x2,x3 is the output power of the No. 1, no. 2 and No. 3 diesel generators respectively; x 4,x5 is the output power of the photovoltaic cell and the energy storage system respectively; x 6 is the amount of electricity that the microgrid purchases or sells to the main grid.
The economic index optimization model is:
Wherein C f(xi) is the fuel cost of the ith power supply; c om(xi) is the maintenance cost of the ith power supply; c grid(x6) is the cost of purchasing electricity from the main grid or the benefit of selling electricity by the micro-grid.
The environment-friendly index optimization model is as follows:
where E (x i) is the pollution emission of the ith power supply.
The safety index optimization model is as follows:
f3(x)＝[Ploss(x)-Ploss,ref]2
wherein P loss (x) is the actual load loss of the micro-grid; p loss,ref is the reference load loss of the microgrid.
Time-invariant constraints include:
Wherein x i is the output power of the ith power supply; x i,min is the minimum output power of the ith power supply; x i,max is the maximum output power of the ith power supply; SOC is the state of charge of the energy storage system; SOC min is the maximum allowed value of state of charge; SOC max is the minimum allowed for state of charge; p dischar,max is the maximum value of the discharge power of the energy storage system; p char,max is the energy storage system discharge power minimum.
The time-varying constraints include:
where P load(tk) optimizes the microgrid load demand for the mission during sub-period t k.
In some embodiments of the invention, the variable parameter set is:
H(t)＝{H(t1),H(t2),...,H(tk),...,H(tK)}
H(tk)＝{H1(tk),H2(tk),...,Hr(tk),...,HR(tk)}
Hr(tk)＝{hr(tk,1),hr(tk,2),...,hr(tk,l),...,hr(tk,L)}
Wherein H (t) is a variable parameter set; h (t k) is a subset of variable parameters within the task optimization subcycle t k; k is the number of variable parameter subsets; h r(tk) is a set of the r-th variable parameters within the task optimization sub-period t k; h r(tk,l) is the parameter value of the r-th variable parameter in the l-th task optimization sub-period t k; l is the number of sub-periods for the same task optimization.
In a specific embodiment of the present invention, the optimization subcycle duration is set to 1 hour, and there are 24 task subcycles in total, and each task subcycle includes five types of variable parameters, namely r=5, which are respectively sensitive load power data, non-sensitive load power data, real-time electricity price data, illumination data and temperature data.
Then, as shown in fig. 2, consider 365 days of historical data, i.e., the total number of identical optimization subcycles is 365, and therefore, the variable parameter subset is:
H(tk)＝{H1(tk),H2(tk),H3(tk),H4(tk),H5(tk),H6(tk)}
for the class r variable parameters, there are:
Hr(tk)＝{hr(tk,1),hr(tk,2),...,hr(tk,l),...,hr(tk,365)}
in some embodiments of the present invention, the sensitivity in step S103 is:
Wherein Δo is sensitivity; o max is the maximum value of the resource optimization target model in the offline multi-target optimization process of the resource scheduling model, and O min is the minimum value of the resource optimization target model in the offline multi-target optimization process of the resource scheduling model.
Specifically, as shown in fig. 3, simplifying the resource scheduling model according to sensitivity in step S104, to obtain a simplified resource scheduling model includes:
s301, judging whether the sensitivity is smaller than the threshold sensitivity;
S302, if the sensitivity is smaller than the threshold sensitivity, the resource optimization target model corresponding to the sensitivity is a resource optimization target model to be deleted, and the resource optimization target model to be deleted is deleted to obtain a simplified resource scheduling model;
s303, if the sensitivity is greater than or equal to the threshold sensitivity, the resource scheduling model is a simplified resource scheduling model.
It should be understood that: the threshold sensitivity can be set or adjusted according to the actual application scene. In a specific embodiment of the present invention, when the resource scheduling model is applied to the micro-grid, the threshold sensitivities of the economic index optimization model, the environmental protection index optimization model, and the safety index optimization model are different, specifically: the threshold sensitivity of the economic index optimization model and the threshold sensitivity of the environmental protection index optimization model are both 0.001, and the threshold sensitivity of the safety index optimization model is 0.01.
In order to avoid the early algorithm maturing due to too many choices of reference vectors and the reduced accuracy of the pareto solution set, or the reduced efficiency of the pareto solution set due to too few choices of reference vectors when the simplified resource scheduling model is optimized in real time, in some embodiments of the present invention, as shown in fig. 4, after step S104, the method further includes:
s401, acquiring an actual optimization sub-period, and calling a simplified resource scheduling model and a reference vector from an offline knowledge base based on the real-time optimization sub-period;
s402, determining a target reference vector from the reference vectors based on a vector determination model;
S403, performing real-time multi-objective optimization on the simplified resource scheduling model based on the evolutionary multi-objective optimization algorithm and the objective reference vector, determining a real-time pareto optimal solution set, and performing resource scheduling based on the real-time pareto optimal solution set.
According to the embodiment of the invention, the target reference vector is determined from the reference vectors based on the vector determination model, so that the efficiency and accuracy of the obtained real-time pareto optimal solution set are improved.
In some embodiments of the present invention, as shown in fig. 5, step S402 includes:
S501, determining the number of target reference vectors in a model from the reference vectors based on the vectors;
s502, randomly determining target reference vectors from the reference vectors based on the number of the target reference vectors.
Specifically, the vector determination model in step S501 is:
Wherein S ref is the number of target reference vectors; g max is the maximum number of iterations; g current is the current iteration number; To round operators.
In some embodiments of the present invention, after step S403, further comprising:
and storing the real-time pareto optimal solution set as a real-time reference vector into an offline knowledge base.
According to the embodiment of the invention, the real-time pareto optimal solution set is stored as the real-time reference vector into the offline knowledge base, so that the reference vector in the offline knowledge base can be expanded, and the diversity of the obtained real-time pareto optimal solution set can be improved.
In summary, the method for constructing the resource scheduling model based on dynamic multi-objective optimization provided by the embodiment of the invention constructs the resource scheduling model before the real-time multi-objective resource scheduling task, and stores the simplified resource scheduling model and the reference vector obtained by offline calculation into an offline knowledge base. Because the calculation link adopts an offline mode, more calculation resources and enough running time can be utilized to obtain a more reasonable simplified resource scheduling model and reference vectors. In the process of real-time multi-target resource scheduling task, the simplified resource scheduling model and the reference vector in the offline knowledge base are called based on the real-time optimization subcycle, so that the complexity of multi-target optimization task calculation can be reduced to a certain extent, and the calculation efficiency is improved by means of fast convergence of the reference vector to a feasible solution area. In addition, the real-time pareto optimal solution set obtained by solving is stored as a real-time reference vector into an offline knowledge base and is used for expanding the reference vector base to realize dynamic update of the offline knowledge base, so that more reference vectors are provided for a later-stage real-time optimization task, the diversity of the population is further improved, the complete pareto solution set is solved, and the accuracy of multi-objective optimization is improved.
In order to better implement the resource scheduling model construction method based on the dynamic multi-objective optimization in the embodiment of the present invention, correspondingly, the embodiment of the present invention further provides a resource scheduling model construction system based on the dynamic multi-objective optimization, as shown in fig. 6, a resource scheduling model construction system 600 based on the dynamic multi-objective optimization includes:
The resource scheduling model construction module 601 is configured to determine a task optimization total period, a plurality of task optimization subcycles, a plurality of resource optimization target models, a decision variable, a time-invariant constraint condition and a time-variant constraint condition of a resource scheduling task, and construct a resource scheduling model based on the task optimization total period, the plurality of task optimization subcycles, the plurality of optimization target models, the decision variable, the time-invariant constraint condition and the time-variant constraint condition;
a variable parameter set determining module 602, configured to determine variable parameters affecting the resource scheduling model based on the historical data, and divide the variable parameters based on a task optimization total period and a plurality of task optimization sub-periods to generate a variable parameter set, where the variable parameter set includes a plurality of variable parameter subsets corresponding to the plurality of task optimization sub-periods one to one;
The multi-objective optimization module 603 is configured to perform offline multi-objective optimization on the resource scheduling model based on an evolutionary multi-objective optimization algorithm and a plurality of variable parameter subsets, and determine a pareto optimal solution set of each task optimization sub-period and sensitivity of each resource optimization objective model in the plurality of resource optimization objective models to a decision variable;
The simplified resource scheduling model construction module 604 is configured to store the pareto optimal solution set as a reference vector to an offline knowledge base, simplify the resource scheduling model according to sensitivity, obtain a simplified resource scheduling model, and store the simplified resource scheduling model to the offline knowledge base.
The resource scheduling model building system 600 based on dynamic multi-objective optimization provided in the foregoing embodiment may implement the technical solution described in the foregoing embodiment of the resource scheduling model building method based on dynamic multi-objective optimization, and the specific implementation principle of each module or unit may refer to the corresponding content in the foregoing embodiment of the resource scheduling model building method based on dynamic multi-objective optimization, which is not described herein again.
As shown in fig. 7, the present invention further provides an electronic device 700 accordingly. The electronic device 700 includes a processor 701, a memory 702, and a display 703. Fig. 7 shows only some of the components of the electronic device 700, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead.
The processor 701 may be, in some embodiments, a central processing unit (Central Processing Unit, CPU), microprocessor or other data processing chip for executing program code or processing data stored in the memory 702, such as the dynamic multi-objective optimization-based resource scheduling model construction method of the present invention.
In some embodiments, the processor 701 may be a single server or a group of servers. The server farm may be centralized or distributed. In some embodiments, the processor 701 may be local or remote. In some embodiments, the processor 701 may be implemented in a cloud platform. In an embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an inter-internal, multiple clouds, or the like, or any combination thereof.
The memory 702 may be an internal storage unit of the electronic device 700 in some embodiments, such as a hard disk or memory of the electronic device 700. The memory 702 may also be an external storage device of the electronic device 700 in other embodiments, such as a plug-in hard disk provided on the electronic device 700, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD), etc.
Further, the memory 702 may also include both internal storage units and external storage devices of the electronic device 700. The memory 702 is used for storing application software and various types of data for installing the electronic device 700.
The display 703 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like in some embodiments. The display 703 is used for displaying information on the electronic device 700 and for displaying a visual user interface. The components 701-703 of the electronic device 700 communicate with each other over a system bus.
In one embodiment, when the processor 701 executes the resource scheduling model building program based on dynamic multi-objective optimization in the memory 702, the following steps may be implemented:
determining a task optimization total period, a plurality of task optimization subcycles, a plurality of resource optimization target models, decision variables, time-invariant constraint conditions and time-variant constraint conditions of a resource scheduling task, and constructing a resource scheduling model based on the task optimization total period, the plurality of task optimization subcycles, the plurality of optimization target models, the decision variables, the time-invariant constraint conditions and the time-variant constraint conditions;
Determining variable parameters affecting the resource scheduling model based on historical data, dividing the variable parameters based on the task optimization total period and the plurality of task optimization subcycles, and generating a variable parameter set, wherein the variable parameter set comprises a plurality of variable parameter subsets which are in one-to-one correspondence with the plurality of task optimization subcycles;
Performing off-line multi-objective optimization on the resource scheduling model based on an evolutionary multi-objective optimization algorithm and the variable parameter subsets, and determining the pareto optimal scheme set of each task optimization sub-period and the sensitivity of each resource optimization target model in the resource optimization target models to the decision variables;
and storing the pareto optimal scheme set as a reference vector to an offline knowledge base, simplifying the resource scheduling model according to the sensitivity to obtain a simplified resource scheduling model, and storing the simplified resource scheduling model to the offline knowledge base.
It should be understood that: the processor 701 may perform other functions in addition to the above functions when executing the resource scheduling model building program based on dynamic multi-objective optimization in the memory 702, see in particular the description of the corresponding method embodiments above.
Further, the type of the electronic device 700 is not particularly limited, and the electronic device 700 may be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (personal digitalassistant, PDA), a wearable device, a laptop (laptop), etc. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices that carry IOS, android, microsoft or other operating systems. The portable electronic device described above may also be other portable electronic devices, such as a laptop computer (laptop) or the like having a touch-sensitive surface, e.g. a touch panel. It should also be appreciated that in other embodiments of the invention, the electronic device 700 may not be a portable electronic device, but rather a desktop computer having a touch-sensitive surface (e.g., a touch panel).
Correspondingly, the embodiment of the application also provides a computer readable storage medium, which is used for storing a computer readable program or instruction, and when the program or instruction is executed by a processor, the steps or functions in the resource scheduling model building method based on dynamic multi-objective optimization provided by the above method embodiments can be realized.
Those skilled in the art will appreciate that all or part of the flow of the methods of the embodiments described above may be accomplished by way of a computer program stored in a computer readable storage medium to instruct related hardware (e.g., a processor, a controller, etc.). The computer readable storage medium is a magnetic disk, an optical disk, a read-only memory or a random access memory.
The method and system for constructing the resource scheduling model based on dynamic multi-objective optimization provided by the invention are described in detail, and specific examples are applied to the description of the principle and the implementation mode of the invention, and the description of the examples is only used for helping to understand the method and the core idea of the invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present invention, the present description should not be construed as limiting the present invention.

Claims (10)

1. The method for constructing the resource scheduling model based on dynamic multi-objective optimization is characterized by comprising the following steps of:
determining a task optimization total period, a plurality of task optimization subcycles, a plurality of resource optimization target models, decision variables, time-invariant constraint conditions and time-variant constraint conditions of a resource scheduling task, and constructing a resource scheduling model based on the task optimization total period, the plurality of task optimization subcycles, the plurality of optimization target models, the decision variables, the time-invariant constraint conditions and the time-variant constraint conditions;
Determining variable parameters affecting the resource scheduling model based on historical data, dividing the variable parameters based on the task optimization total period and the plurality of task optimization subcycles, and generating a variable parameter set, wherein the variable parameter set comprises a plurality of variable parameter subsets which are in one-to-one correspondence with the plurality of task optimization subcycles;
Performing off-line multi-objective optimization on the resource scheduling model based on an evolutionary multi-objective optimization algorithm and the variable parameter subsets, and determining the pareto optimal scheme set of each task optimization sub-period and the sensitivity of each resource optimization target model in the resource optimization target models to the decision variables;
and storing the pareto optimal scheme set as a reference vector to an offline knowledge base, simplifying the resource scheduling model according to the sensitivity to obtain a simplified resource scheduling model, and storing the simplified resource scheduling model to the offline knowledge base.
2. The method for constructing a resource scheduling model based on dynamic multi-objective optimization according to claim 1, wherein the time-invariant constraint conditions include a time-invariant equality constraint condition and a time-invariant inequality constraint condition, and the time-variant constraint conditions include a time-variant equality constraint condition and a time-variant inequality constraint condition; the resource scheduling model is as follows:
minF(x,T)＝{fa(x),a＝1,2,...,N}
x＝(xm,m＝1,2,...,M)∈X
Wherein, minF (x, T) is a resource scheduling model; f a (x) is a optimization target models; n is the total number of the plurality of optimization target models; x is a decision vector; the T task is optimizing the total period duration; t is the time length of the task optimization subcycle; x m is the mth decision variable; m is the total number of decision variables; Is the i 1 th time-invariant inequality constraint condition; /(I) A constraint for the j 1 th time-invariant equation; /(I)Is the i 2 th time-varying inequality constraint; /(I)A 2 th time-varying equality constraint; p 1 is the total number of time-invariant inequality constraints; q 1 is the total number of time-invariant equation constraints; p 2 is the total number of time-varying inequality constraints; q 2 is the total number of time-varying equation constraints; x is the feasible region.
3. The method for constructing a resource scheduling model based on dynamic multi-objective optimization according to claim 1, wherein the variable parameter set is:
H(t)＝{H(t1),H(t2),...,H(tk),...,H(tK)}
H(tk)＝{H1(tk),H2(tk),...,Hr(tk),...,HR(tk)}
Hr(tk)＝{hr(tk,1),hr(tk,2),...,hr(tk,l),...,hr(tk,L)}
Wherein H (t) is a variable parameter set; h (t k) is a subset of variable parameters within the task optimization subcycle t k; k is the number of variable parameter subsets; h r(tk) is a set of the r-th variable parameters within the task optimization sub-period t k; h r(tk,l) is the parameter value of the r-th variable parameter in the l-th task optimization sub-period t k; l is the number of sub-periods for the same task optimization.
4. The method for constructing a resource scheduling model based on dynamic multi-objective optimization according to claim 1, wherein the sensitivity is:
Wherein Δo is sensitivity; o max is the maximum value of the resource optimization target model in the offline multi-target optimization process of the resource scheduling model, and O min is the minimum value of the resource optimization target model in the offline multi-target optimization process of the resource scheduling model.
5. The method for constructing a resource scheduling model based on dynamic multi-objective optimization according to claim 4, wherein the simplifying the resource scheduling model according to the sensitivity to obtain a simplified resource scheduling model comprises:
Judging whether the sensitivity is smaller than a threshold sensitivity or not;
If the sensitivity is smaller than the threshold sensitivity, the resource optimization target model corresponding to the sensitivity is a resource optimization target model to be deleted, and the resource optimization target model to be deleted is deleted to obtain the simplified resource scheduling model;
And if the sensitivity is greater than or equal to the threshold sensitivity, the resource scheduling model is the simplified resource scheduling model.
6. The method for constructing a resource scheduling model based on dynamic multi-objective optimization according to claim 1, further comprising:
Acquiring a real-time optimization sub-period, and calling the simplified resource scheduling model and the reference vector from the offline knowledge base based on the real-time optimization sub-period;
determining a target reference vector from the reference vectors based on a vector determination model;
And carrying out real-time multi-objective optimization on the simplified resource scheduling model based on the evolutionary multi-objective optimization algorithm and the objective reference vector, determining a real-time pareto optimal solution set, and carrying out resource scheduling based on the real-time pareto optimal solution set.
7. The method for constructing a dynamic multi-objective optimization-based resource scheduling model according to claim 6, wherein the determining a target reference vector from the reference vectors based on the vector determination model comprises:
determining a number of target reference vectors of a model from the reference vectors based on the vectors;
The target reference vector is randomly determined from the reference vectors based on the number of target reference vectors.
8. The method for constructing a resource scheduling model based on dynamic multi-objective optimization according to claim 6, further comprising:
And storing the real-time pareto optimal solution set as a real-time reference vector to the offline knowledge base.
9. The method for constructing a resource scheduling model based on dynamic multi-objective optimization according to claim 7, wherein the vector determination model is:
Wherein S ref is the number of target reference vectors; g max is the maximum number of iterations; g current is the current iteration number; To round operators.
10. A resource scheduling model building system based on dynamic multi-objective optimization, comprising:
The resource scheduling model construction module is used for determining a task optimization total period, a plurality of task optimization subcycles, a plurality of resource optimization target models, decision variables, time-invariant constraint conditions and time-variant constraint conditions of a resource scheduling task and constructing a resource scheduling model based on the task optimization total period, the plurality of task optimization subcycles, the plurality of optimization target models, the decision variables, the time-invariant constraint conditions and the time-variant constraint conditions;
A variable parameter set determining module, configured to determine variable parameters affecting the resource scheduling model based on historical data, and divide the variable parameters based on the task optimization total period and the plurality of task optimization sub-periods, to generate a variable parameter set, where the variable parameter set includes a plurality of variable parameter subsets corresponding to the plurality of task optimization sub-periods one to one;
The multi-objective optimization module is used for carrying out off-line multi-objective optimization on the resource scheduling model based on an evolutionary multi-objective optimization algorithm and the variable parameter subsets, and determining the pareto optimal scheme set of each task optimization sub-period and the sensitivity of each resource optimization target model in the resource optimization target models to the decision variables;
And the simplified resource scheduling model construction module is used for storing the pareto optimal scheme set as a reference vector to an offline knowledge base, simplifying the resource scheduling model according to the sensitivity to obtain a simplified resource scheduling model, and storing the simplified resource scheduling model to the offline knowledge base.
CN202210645045.9A 2022-06-06 2022-06-06 Resource scheduling model construction method and system based on dynamic multi-objective optimization Active CN114997659B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210645045.9A CN114997659B (en) 2022-06-06 2022-06-06 Resource scheduling model construction method and system based on dynamic multi-objective optimization

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210645045.9A CN114997659B (en) 2022-06-06 2022-06-06 Resource scheduling model construction method and system based on dynamic multi-objective optimization

Publications (2)

Publication Number Publication Date
CN114997659A CN114997659A (en) 2022-09-02
CN114997659B true CN114997659B (en) 2024-06-14

Family

ID=83033717

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210645045.9A Active CN114997659B (en) 2022-06-06 2022-06-06 Resource scheduling model construction method and system based on dynamic multi-objective optimization

Country Status (1)

Country Link
CN (1) CN114997659B (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107492892A (en) * 2017-08-31 2017-12-19 重庆邮电大学 A kind of multiple target active power dispatch method for solving and system

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111582531A (en) * 2019-02-19 2020-08-25 顺丰科技有限公司 Model creation method, distribution optimization method, device, equipment and storage medium
CN112333712B (en) * 2019-08-05 2022-08-23 中国移动通信集团设计院有限公司 Network planning resource processing method and device
CN111260129B (en) * 2020-01-15 2023-04-07 深圳大学 Multi-yard vehicle path planning method and device, computer equipment and storage medium

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107492892A (en) * 2017-08-31 2017-12-19 重庆邮电大学 A kind of multiple target active power dispatch method for solving and system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
基于MOEA的微电网多目标多约束优化调度研究;李昕;中国博士学位论文全文数据库工程科技Ⅱ辑;20200115;全文 *

Also Published As

Publication number Publication date
CN114997659A (en) 2022-09-02

Similar Documents

Publication Publication Date Title
Yuan et al. A new methodology for optimal location and sizing of battery energy storage system in distribution networks for loss reduction
Li et al. Incorporating energy storage and user experience in isolated microgrid dispatch using a multi‐objective model
Xu et al. Online learning for offloading and autoscaling in renewable-powered mobile edge computing
Moazami Goodarzi et al. An optimal autonomous microgrid cluster based on distributed generation droop parameter optimization and renewable energy sources using an improved grey wolf optimizer
Yin et al. A secured social-economic framework based on PEM-blockchain for optimal scheduling of reconfigurable interconnected microgrids
Premkumar et al. An efficient and reliable scheduling algorithm for unit commitment scheme in microgrid systems using enhanced mixed integer particle swarm optimizer considering uncertainties
CN110705863A (en) Energy optimization scheduling device, equipment and medium
Wang et al. Robust multi-objective load dispatch in microgrid involving unstable renewable generation
Mohseni et al. Stochastic, resilience-oriented optimal sizing of off-grid microgrids considering EV-charging demand response: An efficiency comparison of state-of-the-art metaheuristics
JP2016040997A (en) Energy management system, power supply and demand plan optimization method, and power supply and demand plan optimization program
CN114221340A (en) Distribution network method, device, equipment and medium based on source network load storage distributed energy
Liu et al. Chance‐constrained scheduling model of grid‐connected microgrid based on probabilistic and robust optimisation
US20110295635A1 (en) Systems and methods for scheduling power sources and jobs in an integrated power system
CN104184172B (en) A kind of microgrid energy control method, GPU processors and system
CN114997659B (en) Resource scheduling model construction method and system based on dynamic multi-objective optimization
Zhou et al. Learning-based green workload placement for energy internet in smart cities
Dhawale et al. Optimum generation scheduling incorporating wind energy using HHO–IGWO algorithm
Xie et al. Analysis of controllable capacity for electric vehicle battery swapping stations
CN111525629B (en) Power supply capacity configuration method and device, computer equipment and storage medium
CN114123187A (en) Virtual power plant planning method and device, electronic equipment and storage medium
Rizki et al. Economic dispatch problem in smart grid system with considerations for pumped storage
Dieu et al. Enhanced merit order and augmented Lagrange Hopfield network for ramp rate constrained unit commitment
CN111626485A (en) Load prediction system and method for regional building energy system
CN117541030B (en) Virtual power plant optimized operation method, device, equipment and medium
CN118300195B (en) Intelligent monitoring and scheduling method for micro-grid

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