CN-121979649-A - Photovoltaic electric field task scheduling method, system, equipment and medium

CN121979649ACN 121979649 ACN121979649 ACN 121979649ACN-121979649-A

Abstract

The invention discloses a photovoltaic electric field task scheduling method, a system, equipment and a medium, wherein the method comprises the steps of collecting equipment position data of a photovoltaic electric field to construct a photovoltaic electric field equipment communication topology network; and obtaining a pareto front optimal solution set, and obtaining a task unloading decision and a resource allocation strategy of an edge server. The method avoids the problem of high cloud data long-distance return delay by introducing edge calculation, constructs a multi-objective optimization problem by taking the delay performance index and the load balancing performance index as optimization targets, ensures the stability of a queue through scalar function optimization, obtains the pareto front edge optimal solution set through multi-objective evolutionary algorithm optimization, can flexibly adapt to different operation and maintenance scene priorities, effectively reduces task delay, improves the resource utilization rate of an edge server, and enhances the instantaneity and stability of photovoltaic electric field dispatching.

Inventors

GAN RUNDONG
Fan Junqiu
WANG BIN
LUO CHEN
WANG WEI
MOU XUEPENG
WANG CE
YANG SHIPING
CHEN JULONG
ZHANG YU

Assignees

贵州电网有限责任公司

Dates

Publication Date: 20260505
Application Date: 20251128

Claims (10)

1. The photovoltaic electric field task scheduling method is characterized by comprising the following steps of: collecting equipment position data of a photovoltaic electric field, and constructing a photovoltaic electric field equipment communication topology network according to the equipment position data; constructing a task scheduling model, a communication transmission model and a calculation processing model according to the photovoltaic electric field equipment communication topology network; Constructing a multi-objective optimization problem according to the task scheduling model, the communication transmission model and the calculation processing model; queue stability constraint is carried out in the multi-objective optimization problem, and the constrained multi-objective optimization problem is converted into a deterministic optimization problem; And solving the deterministic optimization problem by adopting a multi-objective evolutionary algorithm to obtain a pareto front-edge optimal solution set, and obtaining a task unloading decision and a resource allocation strategy of an edge server from the pareto front-edge optimal solution set according to comprehensive balance of the optimization objective.
2. The photovoltaic electric field task scheduling method according to claim 1, wherein the step of constructing a communication topology network of the photovoltaic electric field device comprises: collecting device position data of devices in a photovoltaic electric field and various devices in an edge server; And constructing a photovoltaic electric field equipment communication topology network according to the equipment position data.
3. The photovoltaic electric field task scheduling method according to claim 2, wherein the steps of constructing a task scheduling model, a communication transmission model, and a calculation processing model include: The method comprises the steps of dividing a communication scheduling period in a photovoltaic electric field equipment communication topological network into a plurality of time slots with equal length, generating task data for each equipment in the time slots, establishing an initial data buffer queue for each equipment to buffer the task data, and obtaining a task scheduling model after setting a queue upper limit constraint on the initial data buffer queue; Constructing a communication transmission model from equipment to an edge server of the task data according to the photovoltaic electric field equipment communication topology network; and processing and unloading task data on the edge server through the task scheduling model, and constructing a calculation processing model.
4. The photovoltaic electric field task scheduling method of claim 3, wherein the step of constructing a multi-objective optimization problem comprises: obtaining a time delay performance index through the communication transmission model and the calculation processing model, obtaining a load balancing performance index through the calculation processing model, and obtaining an optimization target by combining the time delay performance index and the load balancing performance index; obtaining an optimization decision through task unloading of the task scheduling model, a transmission link of the communication transmission model and resource allocation of the calculation processing model; obtaining constraint conditions by combining the queue upper limit constraint in the task scheduling model, the access constraint in the communication transmission model and the capacity constraint in the calculation processing model; And constructing an optimization decision and a constraint condition according to the optimization target to obtain a multi-target optimization problem.
5. The photovoltaic electric field task scheduling method of claim 4, wherein the step of translating into a deterministic optimization problem comprises: Modifying the initial data cache queue in combination with task data according to the upper limit constraint of the queue in the task scheduling model to obtain a modified data cache queue; defining a scalar function according to the corrected data cache queue, combining the optimization targets of the multi-target optimization problem, constructing a drift penalty function, deriving the upper limit of the drift penalty function, and forming a drift penalty analysis result; and converting the multi-objective optimization problem into a deterministic optimization problem without long-term queue constraint by minimizing the drift penalty analysis result.
6. The method for photovoltaic electric field task scheduling according to claim 5, wherein the step of obtaining the pareto front optimal solution set comprises: According to the deterministic optimization problem, combining task unloading, transmission links and resource allocation in the optimization decision to encode to obtain an initial solution of the multi-objective evolutionary algorithm, and according to the initial solution, configuring the operation parameters of the multi-objective evolutionary algorithm to generate an initial population; Screening the initial population by combining the constraint condition to obtain a compliance population, and carrying out iterative updating on the compliance population according to the optimization target to obtain an iterative population; presetting maximum iteration times, and outputting the pareto front optimal solution set when the iteration times of the iteration population reach the maximum iteration times.
7. The photovoltaic electric field task scheduling method according to claim 6, wherein the step of obtaining a task offloading decision and a resource allocation policy of the edge server comprises: Presetting weights of a time delay performance index and a load balancing performance index in the optimization target, carrying out comprehensive cost evaluation on the pareto front optimal solution set, and calculating to obtain a comprehensive cost value; and selecting a solution with the optimal comprehensive cost value as a target solution, and analyzing the target solution to obtain a task unloading decision and a resource allocation strategy of the edge server.
8. A photovoltaic electric field task scheduling system applying the method of any one of claims 1-7, comprising: the data acquisition module is used for acquiring equipment position data of equipment and an edge server in the photovoltaic electric field; The network construction module is used for constructing a photovoltaic electric field equipment communication topology network according to the equipment position data; the model construction module is used for constructing a task scheduling model, a communication transmission model and a calculation processing model according to the photovoltaic electric field equipment communication topology network; The optimization problem construction module is used for constructing a multi-objective optimization problem by combining the task scheduling model, the communication transmission model and the calculation processing model; The problem conversion module is used for introducing queue stability constraint to the multi-objective optimization problem and converting the multi-objective optimization problem into a deterministic optimization problem; the solving module is used for solving the deterministic optimization problem by adopting a multi-objective evolutionary algorithm and outputting a pareto front edge optimal solution set; and the decision generation module is used for generating a task unloading decision and a resource allocation strategy of the edge server according to the pareto front optimal solution set and the weight evaluation of the optimization target.
9. An electronic device, comprising: A memory and a processor; The memory is configured to store computer executable instructions that, when executed by the processor, implement the steps of the photovoltaic electric field task scheduling method of any one of claims 1 to 7.
10. A computer readable storage medium storing computer executable instructions which when executed by a processor implement the steps of the photovoltaic electric field task scheduling method of any one of claims 1 to 7.

Description

Photovoltaic electric field task scheduling method, system, equipment and medium Technical Field The invention relates to the technical field of edge calculation, in particular to a photovoltaic electric field task scheduling method, a system, equipment and a medium. Background The photovoltaic electric field is used as a core infrastructure for new energy consumption and low-carbon transformation of a power grid, and needs to continuously collect massive environment and equipment data such as solar radiation, temperature, equipment current and voltage, inverter running state and the like, so that the photovoltaic electric field is used for short-time power generation output prediction, fault early warning positioning and operation and maintenance scheduling. The real-time processing of the data directly affects grid-connected safety and power generation income, if the data processing is lagged, not only the output prediction precision is reduced, the power generation plan deviation is caused, but also the fault disposal can be delayed, and the economic loss and the safety risk are caused. With the gradual application of the edge computing technology in the energy field, the near-end data processing is realized by deploying an edge server in a photovoltaic electric field, so that a data return link is shortened, and the real-time response capability is improved, and the method has become an important development direction of the task scheduling of the photovoltaic electric field. At present, the traditional cloud centralized processing needs to transmit original data back to a cloud center for a long distance, is easily influenced by factors such as uplink and downlink bandwidth limitation, link congestion and the like, has high end-to-end time delay and large jitter, is easy to form a processing bottleneck in a sudden alarm and high concurrency acquisition scene, is difficult to meet the near real-time analysis and linkage control requirements of a station side, and a partial station side scheduling scheme focuses on a single optimization target only, if only pursues minimum time delay or maximum throughput, does not comprehensively consider the load balance of a plurality of edge servers, leads to the task being excessively concentrated on a few edge servers with better performance, causes the problems of long-term overload, queue backlog, task overtime discarding and the like, and other servers are in idle state, so that the overall resource utilization rate is reduced, the service stability is insufficient, and the task processing requirements of a photovoltaic electric field are not suitable for being complicated and changeable. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides a photovoltaic electric field task scheduling method, system, equipment and medium, which solve the problems that the traditional cloud centralized processing needs long-distance data returning, is easily influenced by bandwidth and congestion, has high time delay and large jitter, is difficult to meet near real-time requirements, and a partial station side scheme focuses on only a single optimization target, does not overall the load balance of an edge server, and causes the problems of concentrated task, resource waste and poor service stability. In order to solve the technical problems, the invention provides the following technical scheme: in a first aspect, the present invention provides a photovoltaic electric field task scheduling method, including: collecting equipment position data of a photovoltaic electric field, and constructing a photovoltaic electric field equipment communication topology network according to the equipment position data; constructing a task scheduling model, a communication transmission model and a calculation processing model according to the photovoltaic electric field equipment communication topology network; Constructing a multi-objective optimization problem according to the task scheduling model, the communication transmission model and the calculation processing model; queue stability constraint is carried out in the multi-objective optimization problem, and the constrained multi-objective optimization problem is converted into a deterministic optimization problem; And solving the deterministic optimization problem by adopting a multi-objective evolutionary algorithm to obtain a pareto front-edge optimal solution set, and obtaining a task unloading decision and a resource allocation strategy of an edge server from the pareto front-edge optimal solution set according to comprehensive balance of the optimization objective. As a preferable scheme of the photovoltaic electric field task scheduling method, the method comprises the following steps of: collecting device position data of devices in a photovoltaic electric field and various devices in an edge server; And constructing a phot