Layered virtual power plant transaction optimization scheduling method

Abstract

The invention discloses a layered virtual power plant transaction optimization scheduling method, and belongs to the technical field of virtual power plant optimization. The method comprises the following steps of preprocessing working data of a virtual power plant, analyzing a transaction saturation index of a server, dividing a server transfer receiving state according to the transaction saturation index, analyzing a conforming index and optimizing a scheduling task according to the conforming index. According to the method, the matching index between the server corresponding to the transferable index of the transaction task and the server corresponding to the receivable index of the transaction task is calculated, the probability of task interaction between the servers is evaluated, and the task allocation between the virtual power plant servers is adjusted according to the probability, so that the computing resources of each server are fully utilized; the situation that some servers are idle and resources of other servers are exhausted is avoided, and the problems that in the prior art, loads of virtual power plant servers are unbalanced and resource distribution is insufficient are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plant optimization, and in particular to a layered virtual power plant transaction optimization scheduling method. Background Art

[0002] In recent years, the clean energy production, electrification of energy consumption, and high efficiency of energy utilization have shown an accelerated development trend. The operation of the new power system has shown the characteristics of "double highs and double peaks", and the demand for flexible power supply adjustment has become increasingly urgent. As the fusion point of energy supply and energy consumption, the virtual power plant (VPP) uses advanced information and communication technology to aggregate distributed resources such as distributed power sources, electric vehicles, heat storage (cold storage), electric energy storage, and hydrogen energy storage. It realizes the coordinated and optimized operation of distributed resources through a higher-level software architecture. The lower layer of the virtual power plant is mainly the virtual power plant and the aggregator or resource entity, and the upper layer is mainly the virtual power plant and the electric energy market entity.

[0003] At present, hierarchical virtual power plant scheduling has achieved scheduling optimization for different levels through multi-time scale optimization methods (including day-ahead optimization, intraday rolling optimization and real-time adjustment stages) and distributed model predictive control algorithms, combined with the hierarchical distributed optimization model of the upper virtual power plant and the lower micro energy network.

[0004] For example, the invention patent with publication number: CN115829248A discloses a virtual power plant multi-time scale hierarchical optimization scheduling method and system, and the scheduling method includes: day-ahead optimization stage, intraday rolling optimization stage and real-time adjustment stage. By optimizing the call of existing adjustable resources to ensure the safe and stable operation of the system and save the investment in grid-side equipment, the use of multi-time scale optimization scheduling can reduce the impact of the uncertainty of outdoor meteorological parameters and manually set parameters on the operation of the virtual power plant; based on the distributed model predictive control algorithm, the traditional model predictive control model problem is decomposed into each virtual machine group.

[0005] For example, the invention patent with publication number: CN115392585A discloses a virtual power plant hierarchical distributed optimization scheduling method, device and equipment, including: establishing an upper-layer virtual power plant distributed optimization model and a lower-layer micro-energy network centralized optimization model, solving the upper-layer virtual power plant distributed optimization model and the lower-layer micro-energy network centralized optimization model according to the virtual power plant hierarchical distributed optimization algorithm, and obtaining the optimal power of each distributed power source in the micro-energy network, as well as the minimum comprehensive cost of the micro-energy network and the maximum benefit obtained by the virtual power plant participating in the transaction.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems:

[0007] In the prior art, as the scale of virtual power plants expands, the number of virtual power plant servers and the complexity of tasks gradually increase, causing some servers to be overloaded due to over-intensive tasks or insufficient processing capacity. Overload will further cause task delays or inability to complete, affecting the overall performance and service quality of the system. In summary, there are problems of unbalanced load and insufficient resource allocation on virtual power plant servers. Summary of the invention

[0008] In view of the above-mentioned problems, the present invention is proposed.

[0009] Therefore, the problem to be solved by the present invention is that in the prior art, some servers are overloaded due to overly intensive tasks or insufficient processing capacity, resulting in task delays or inability to complete, and there are problems of unbalanced load and insufficient resource allocation on virtual power plant servers.

[0010] In order to solve the above technical problems, the present invention provides the following technical solutions: a hierarchical virtual power plant transaction optimization scheduling method, which includes the following steps: the virtual power plant control center obtains the virtual power plant working data and pre-processes the virtual power plant working data; based on the pre-processed virtual power plant working data, the transaction saturation index of each virtual power plant corresponding server is analyzed, and the transaction saturation index is used to reflect the saturation degree of transaction task processing in the virtual power plant server; when the transaction saturation index is higher than the first saturation threshold in the database, the transaction task transferability index of the virtual power plant corresponding server is analyzed, and the transaction task transferability index is used to reflect the probability that a transaction task can be transferred to other virtual power plant servers; when the transaction saturation index is higher than the first saturation threshold in the database, ... When the saturation index is not higher than the second saturation threshold in the database, the transaction task receivable index of the corresponding server of the virtual power plant is analyzed, and the transaction task receivable index is used to reflect the probability that a transaction task can be received by the current virtual power plant server; when the transaction saturation index is not higher than the first saturation threshold and higher than the second saturation threshold, no processing is performed on the virtual power plant server corresponding to the transaction saturation index; according to the transaction task transferability index and the transaction task receivable index, the compatibility index of the two virtual power plant corresponding servers is analyzed, and the compatibility index is used to reflect the probability that the two virtual power plant corresponding servers can interact with each other in transaction tasks; the transaction tasks between the virtual power plant servers are adjusted according to the compatibility index.

[0011] As a preferred scheme of the layered virtual power plant transaction optimization scheduling method described in the present invention, wherein: the virtual power plant control center serves as the upper layer of the virtual power plant, and the virtual power plant control center connects and manages the servers corresponding to each virtual power plant; the server serves as the middle layer of the virtual power plant, and the server corresponds to the virtual power plant one-to-one; the virtual power plant serves as the lower layer of the virtual power plant, including various power equipment; the virtual power plant control center and each virtual power plant server exchange data through a communication protocol; in the virtual power plant control center, the virtual power plant working data from the virtual power plant server is obtained and summarized; the virtual power plant control center performs centralized scheduling according to the status and load conditions of different virtual power plant servers; each virtual power plant server collects data on the power equipment it includes.

[0012] As a preferred scheme of the layered virtual power plant transaction optimization scheduling method described in the present invention, the specific method of obtaining the transaction saturation index is: based on the preprocessed virtual power plant working data, the number of tasks, task complexity, response time, and task execution time of the virtual power plant server are extracted; according to the objective weighting method, the weights of the transaction saturation index of the number of tasks, task complexity, response time, and task execution time are obtained respectively; based on the number of tasks, task complexity, response time, task execution time and their weights, the transaction saturation index is obtained through the transaction saturation index formula.

[0013] As a preferred solution of the layered virtual power plant transaction optimization scheduling method of the present invention, the transaction saturation index formula is:

[0014]

[0015] In the formula, SD i is the transaction saturation index of the i-th virtual power plant server, i is the number of the virtual power plant server, i = 1, 2, ..., I, I is the total number of virtual power plant servers, e is a natural constant, TC i is the number of tasks of the i-th virtual power plant server, ONN i is the task complexity of the current task of the i-th virtual power plant server, TR i is the response time of the i-th virtual power plant server, TE i is the task execution time of the i-th virtual power plant server, α1 is TC i For SD i The weight of ONN i For SD i The weight of TR i For SD i The weight of α4 is TE i For SD i The weight of .

[0016] As a preferred scheme of the layered virtual power plant transaction optimization scheduling method described in the present invention, the specific method of obtaining the transaction task transferability index is: based on the preprocessed virtual power plant working data, the current load, server remaining capacity and resource availability of the virtual power plant server are extracted; according to the objective weighting method, the weights of the current load, server remaining capacity and resource availability for the transaction task transferability index are obtained respectively; based on the current load, server remaining capacity, resource availability and their weights, the transaction task transferability index is obtained through the transaction task transferability index formula.

[0017] As a preferred solution of the layered virtual power plant transaction optimization scheduling method of the present invention, the transferable index formula of the transaction task is:

[0018] RT i = sinh[β1*TP i +β2*(1-SC i )+β3*(1-CR i )]

[0019] Where RT i is the transferable index of the transaction task of the i-th virtual power plant server, i is the number of the virtual power plant server, i = 1, 2, ..., I, I is the total number of virtual power plant servers, TP i is the current load of the i-th virtual power plant server, SC i is the remaining capacity of the server in the i-th virtual power plant, CR i is the resource availability of the i-th virtual power plant server, β1 is TP i For RT i The weight of , β2 is SC i For RT i The weight of β3 is CR i For RT i The weight of .

[0020] As an optimal solution of the layered virtual power plant transaction optimization scheduling method described in the present invention, the specific method of obtaining the receivable index of the transaction task is as follows: based on the preprocessed virtual power plant working data, the server remaining capacity, current load, average task execution time, and complexity of tasks to be received of the virtual power plant server are extracted; according to the objective weighting method, the weights of the server remaining capacity, current load, average task execution time, and complexity of tasks to be received for the receivable index of the transaction task are obtained respectively; based on the server remaining capacity, current load, average task execution time, complexity of tasks to be received and their weights, the receivable index of the transaction task is obtained through the receivable index formula of the transaction task.

[0021] As a preferred solution of the layered virtual power plant transaction optimization scheduling method of the present invention, the transaction task receivable index formula is:

[0022] RR i =tanh -1 [δ1*(1-TP i )+δ2*SC i +δ3*(1-TA i )+δ4*(1-ON i )]

[0023] In the formula, RR i is the transaction task receivable index of the i-th virtual power plant server, i is the number of the virtual power plant server, i=1,2,...,I, I is the total number of virtual power plant servers, SC i is the remaining capacity of the server of the i-th virtual power plant server, TP i is the current load of the i-th virtual power plant server, TA i is the average execution time of tasks on the i-th virtual power plant server, ON i is the complexity of the tasks to be received by the i-th virtual power plant server, δ1 is TS i For RR i The weight of ,δ2 is SC i For RR i The weight of , δ3 is TA i For RR i The weight of δ4 is ON i For RR i The weight of .

[0024] As a preferred solution of the layered virtual power plant transaction optimization scheduling method described in the present invention, the specific method of obtaining the matching index is: obtaining the transaction task transferability index, the transaction task receivable index, and the server compatibility data, wherein the server compatibility data includes hardware configuration similarity and communication delay; obtaining the weights of the transaction task transferability index, the transaction task receivable index, and the server compatibility data for the matching index by an objective weight method; based on the transaction task transferability index, the transaction task receivable index, the server compatibility data and their corresponding weights, the matching index is obtained by the matching index formula; the matching index formula is:

[0025]

[0026] In the formula, For the i a and the i-th b The matching index of the virtual power plant server, i is the number of the virtual power plant server, i = 1, 2, ..., i a ,...,ib ,...,I, I is the total number of virtual power plant servers, i a is the number of the virtual power plant server to which the task is to be transferred. b is the number of the virtual power plant server to receive the task. For the i a The transaction task transferability index of the virtual power plant server, For the i a The transaction tasks of the virtual power plant server can receive index, For the i a and the i-th b The hardware configuration similarity between the virtual power plant servers, For the i a The communication delay of the virtual power plant servers, For the i b The communication delay of the virtual power plant server, μ1 is for The weight of μ2 is for The weight of μ3 is the server compatible data for The weight of .

[0027] As a preferred scheme of the layered virtual power plant transaction optimization scheduling method described in the present invention, the specific steps of adjusting the transaction tasks between virtual power plant servers according to the matching index are as follows: when the matching index exceeds the matching threshold in the database, two groups of servers with the highest matching index are selected to transfer the tasks to be received to achieve optimized scheduling; when the tasks to be received are received, the real-time status of the server is updated.

[0028] The beneficial effects of the present invention are as follows: the present invention transfers tasks according to the compatibility index between servers, so that the computing resources of each server are fully utilized, avoiding the situation where some servers are idle while the resources of other servers are exhausted, achieving optimized load balancing and reasonable allocation of resources, and effectively solving the problems of unbalanced load and insufficient resource allocation of virtual power plant servers in the prior art.

[0029] The present invention divides the transfer and reception status of the virtual power plant servers according to the transaction saturation index, thereby extracting servers with excessively high loads and servers with excessively low loads, thereby effectively avoiding server overload, improving resource utilization, and achieving smooth operation and efficient scheduling of the virtual power plant.

[0030] The present invention divides the virtual power plant transaction process into three levels: a virtual power plant control center, a virtual power plant server, and power equipment. The virtual power plant control center monitors the status of each virtual power plant server in real time, is able to detect uneven loads or overloads of certain servers, and dynamically migrates tasks or balances loads as needed, thereby preventing overload or waste of resources and improving overall operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0032] Figure 1 This is a flow chart of the layered virtual power plant transaction optimization scheduling method in Example 1. DETAILED DESCRIPTION

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0034] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0035] This embodiment provides a hierarchical virtual power plant transaction optimization scheduling method. The embodiment of the present application solves the problems of unbalanced load and insufficient resource allocation of virtual power plant servers in the prior art by providing a hierarchical virtual power plant transaction optimization scheduling method. First, the virtual power plant work data is obtained and preprocessed, and then the transaction saturation index of each server is analyzed; when the transaction saturation index exceeds the first threshold, the transaction task transferability index of the server is analyzed to evaluate the probability of task transfer to other servers; when the transaction saturation index is lower than the second threshold, the transaction task receivable index of the server is analyzed to evaluate the probability of the current server receiving the task; if the transaction saturation index is between the two thresholds, the server is not adjusted; then, based on the transaction task transferability index and the transaction task receivable index, the fit index between the server corresponding to the transaction task transferability index and the server corresponding to the transaction task receivable index is calculated, the probability of task interaction between them is evaluated, and the task allocation between the virtual power plant servers is adjusted accordingly, thereby optimizing load balancing and reasonable allocation of resources.

[0036] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0037] like Figure 1 As shown, it is a flow chart of the hierarchical virtual power plant transaction optimization scheduling method provided in an embodiment of the present application, and the method includes the following steps: Preprocessing of virtual power plant working data: The virtual power plant control center obtains the virtual power plant working data and preprocesses the virtual power plant working data. The above preprocessing at least includes unit unification and standardization to remove the dimension of all data.

[0038] Analyze the transaction saturation index of the server: Analyze the transaction saturation index of the corresponding server of each virtual power plant based on the preprocessed virtual power plant work data. The transaction saturation index is used to reflect the saturation level of transaction task processing in the virtual power plant server.

[0039] The server transfer reception status is divided according to the transaction saturation index: when the transaction saturation index is higher than the first saturation threshold in the database, the transaction task transferability index of the server corresponding to the virtual power plant is analyzed, and the transaction task transferability index is used to reflect the probability that a transaction task can be transferred to other virtual power plant servers; when the transaction saturation index is not higher than the second saturation threshold in the database, the transaction task receivable index of the server corresponding to the virtual power plant is analyzed, and the transaction task receivable index is used to reflect the probability that a transaction task can be received by the current virtual power plant server; when the transaction saturation index is not higher than the first saturation threshold and higher than the second saturation threshold, no processing is performed on the virtual power plant server corresponding to the transaction saturation index.

[0040] Analyze the compatibility index and optimize the scheduling tasks accordingly: Analyze the compatibility index of the corresponding servers of the two virtual power plants according to the transaction task transferability index and the transaction task receivability index. The compatibility index is used to reflect the probability that the corresponding servers of the two virtual power plants can interact with each other in transaction tasks; adjust the transaction tasks between the virtual power plant servers according to the compatibility index.

[0041] Furthermore, the virtual power plant control center serves as the upper layer of the virtual power plant, and the virtual power plant control center connects and manages the servers corresponding to each virtual power plant; the server serves as the middle layer of the virtual power plant, and the server corresponds to the virtual power plant one-to-one; the virtual power plant serves as the lower layer of the virtual power plant, including various power equipment; the virtual power plant control center and each virtual power plant server exchange data through a communication protocol; in the virtual power plant control center, the virtual power plant working data from the virtual power plant server is obtained and summarized; the virtual power plant control center performs centralized scheduling according to the status and load conditions of different virtual power plant servers; each virtual power plant server collects data on the power equipment it includes.

[0042] In this embodiment, the virtual power plant control center can obtain and summarize work data from each virtual power plant server, such as the number of tasks, execution time, resource usage, etc. By real-time monitoring the status of each virtual power plant server, the virtual power plant control center can detect uneven load or overload of some servers, and thus dynamically migrate tasks or load balance as needed to prevent overload or waste of resources and improve overall operational efficiency.

[0043] Furthermore, the specific method of obtaining the transaction saturation index is as follows: based on the preprocessed virtual power plant working data, the task quantity, task complexity, response time, and task execution time of the virtual power plant server are extracted; according to the objective weighting method, the weights of the task quantity, task complexity, response time, and task execution time for the transaction saturation index are obtained respectively; based on the task quantity, task complexity, response time, task execution time and their weights, the transaction saturation index is obtained through the transaction saturation index formula.

[0044] Furthermore, the transaction saturation index formula is:

[0045]

[0046] In the formula, SD i is the transaction saturation index of the i-th virtual power plant server, i is the number of the virtual power plant server, i = 1, 2, ..., I, I is the total number of virtual power plant servers, e is a natural constant, TC i is the number of tasks of the i-th virtual power plant server, ONN i is the task complexity of the current task of the i-th virtual power plant server, TR i is the response time of the i-th virtual power plant server, TE i is the task execution time of the i-th virtual power plant server, α1 is TC i For SD i The weight of ONN i For SD i The weight of TR i For SD i The weight of α4 is TE i For SD i The weight of .

[0047] In this embodiment, TC i = 15, the number of tasks currently in progress is obtained from the task scheduling platform through the virtual power plant management system, and the number of tasks can also be obtained by querying the current task status table in the database. i=0.82, the task complexity can be determined by calculating the ratio of the computing resources required by the task to the maximum computing resources. For example, if the CPU time required for the task is 3 seconds, and the maximum available CPU time is 4 seconds, then the task complexity is 3 / 4 = 0.75. TR i = 0.02 seconds, measured by measuring the time required for a task to go from request to response, for example, using a network latency monitoring tool (such as a Ping tool) or through the performance monitoring function of a virtual power plant platform. TE i = 12 seconds, through the monitoring system of the virtual power plant server, record the start time and end time of each task, calculate the actual execution time of the task, and can also be obtained through log files or execution tracking records. α1 = 0.3, α2 = 0.3, α3 = 0.2, α4 = 0.2, transaction saturation index SD i The higher the transaction saturation index is and the closer it is to 1, the higher the current load of the server is. On the contrary, the higher the transaction saturation index is and the closer it is to 0, the lower the current load is.

[0048] Furthermore, the specific method of obtaining the transaction task transferability index is as follows: based on the preprocessed virtual power plant working data, the current load, server remaining capacity, and resource availability of the virtual power plant server are extracted; the weights of the current load, server remaining capacity, and resource availability for the transaction task transferability index are obtained according to the objective weighting method; based on the current load, server remaining capacity, resource availability and their weights, the transaction task transferability index is obtained through the transaction task transferability index formula.

[0049] Further, as a preferred solution of the layered virtual power plant transaction optimization scheduling method of the present invention, the transaction task transferability index formula is:

[0050] RT i = sinh[β1*TP i +β2*(1-SC i )+β3*(1-CR i )]

[0051] Where RT i is the transaction task transferability index of the i-th virtual power plant server, i is the number of the virtual power plant server, TP i is the current load of the i-th virtual power plant server, SC i is the remaining capacity of the server in the i-th virtual power plant, CR i is the resource availability of the i-th virtual power plant server, β1 is TP i For RT i The weight of , β2 is SC i For RT i The weight of β3 is CRi For RT i The weight of .

[0052] In this embodiment, TP i The value is 0.86, which can be obtained through the virtual power plant monitoring system or operating system commands, such as the top command to obtain CPU usage and the free command to obtain memory usage (i.e. current load). The range is 0 to 1 (0 means idle and 1 means full load). i The remaining capacity of the server refers to the proportion of currently available resources on the server, including CPU, memory, and storage. This value can be obtained by analyzing the currently used resources of the server and subtracting the occupied resources, or by obtaining real-time resource usage data through system monitoring tools. The range is 0 to 1 (0 means fully occupied and 1 means completely idle). i The value of β1 is 0.7, β2 is 0.2, and β3 is 0.1. The transaction task transferability index RT is used to obtain the number of successful and failed task scheduling through the task scheduling system or operating system log, and the probability that the resources of the virtual power plant server can be successfully scheduled when executing tasks is obtained. For example, calculate how many tasks the system has successfully completed in a certain period of time in the past, and how many tasks have not been scheduled due to insufficient resources, ranging from 0 to 1 (0 means no resources are available, 1 means that resources are fully available). i The higher the transaction task transferability index is, the greater the probability that the pending tasks in the server will be transferred to other servers, and vice versa.

[0053] Furthermore, the specific method of obtaining the receivable index of the transaction task is as follows: based on the preprocessed virtual power plant working data, the server remaining capacity, current load, average task execution time, and complexity of tasks to be received of the virtual power plant server are extracted; according to the objective weighting method, the weights of the server remaining capacity, current load, average task execution time, and complexity of tasks to be received for the receivable index of the transaction task are obtained respectively; based on the server remaining capacity, current load, average task execution time, complexity of tasks to be received and their weights, the receivable index of the transaction task is obtained through the receivable index formula of the transaction task.

[0054] Furthermore, the formula for the receivable index of the transaction task is:

[0055] RR i =tanh -1 [δ1*(1-TP i )+δ2*SC i +δ3*(1-TA i )+δ4*(1-ON i )]

[0056] In the formula, RR iis the transaction task receivable index of the i-th virtual power plant server, i is the number of the virtual power plant server, SC i is the remaining capacity of the server of the i-th virtual power plant server, TP i is the current load of the i-th virtual power plant server, TA i is the average execution time of tasks on the i-th virtual power plant server, ON i is the complexity of the tasks to be received by the i-th virtual power plant server, δ1 is TS i For RR i The weight of ,δ2 is SC i For RR i The weight of , δ3 is TA i For RR i The weight of δ4 is ON i For RR i The weight of .

[0057] In this embodiment, SC i 0.68,TP i 0.26. i The average task execution time is 0.8 seconds. The average task execution time refers to the average time required for the server to process a task. The average task execution time can be obtained by obtaining the execution time of each task through the task scheduling system or operating system log. i is 0.64. The complexity of the task to be received indicates the complexity of the task to be received by the server, which is obtained by the estimated execution time and resource requirements of the task to be received, or the complexity of the task to be received is obtained based on the historical data model or task scheduling system. δ1 is 0.3, δ2 is 0.3, δ3 is 0.2, and δ4 is 0.2. The transaction task receivable index RR i The larger the transaction task receivable index is, the greater the probability that the server can receive the pending tasks is, and vice versa.

[0058] Furthermore, the specific method of obtaining the fit index is as follows: obtaining the transaction task transferability index, the transaction task receivable index, and the server compatibility data, wherein the server compatibility data includes the hardware configuration similarity and the communication delay; obtaining the weights of the transaction task transferability index, the transaction task receivable index, and the server compatibility data for the fit index through the objective weight method; and obtaining the fit index through the fit index formula based on the transaction task transferability index, the transaction task receivable index, the server compatibility data and their corresponding weights; the fit index formula is:

[0059]

[0060] In the formula, For the i a and the i-thb The matching index of the virtual power plant server, i is the number of the virtual power plant server, i = 1, 2, ..., i a ,...,i b ,...,I,i a is the number of the virtual power plant server to which the task is to be transferred. b is the number of the virtual power plant server to receive the task. For the i a The transaction task transferability index of the virtual power plant server, For the i a The transaction tasks of the virtual power plant server can receive index, For the i a and the i-th b The hardware configuration similarity between the virtual power plant servers, For the i a The communication delay of the virtual power plant servers, For the i b The communication delay of the virtual power plant server, μ1 is for The weight of μ2 is for The weight of μ3 is the server compatible data for The weight of .

[0061] In this embodiment, the transaction task transferability index RT i The transaction task can receive the index RR of 0.8379. i is 0.6013, The hardware configuration similarity is 0.82. The hardware configuration similarity is used to measure the similarity of the hardware resources of two virtual power plant servers, such as CPU type, memory, storage type, etc. A high similarity indicates that the efficiency of the two servers in processing the same type of tasks is similar. It can be obtained through Euclidean distance or cosine similarity, and the range is [0, 1] (1 means complete similarity, 0 means complete dissimilarity). is 0.01, The communication delay is 0.02. The communication delay is used to measure the delay of data transmission between two virtual power plant servers. The smaller the delay, the higher the cooperation efficiency between servers. It is measured using network performance monitoring tools, or the delay is estimated by monitoring the performance of the network interface. μ1 is 0.3, μ2 is 0.3, and μ3 is 0.4. Fit index The higher the fit index, the better the effect of task scheduling between the two servers, and vice versa.

[0062] Furthermore, the specific steps for adjusting the transaction tasks between virtual power plant servers according to the matching index are as follows: when the matching index exceeds the matching threshold in the database, two groups of servers with the highest matching index are selected to transfer the tasks to be received to achieve optimized scheduling; when the tasks to be received are received, the real-time status of the server is updated.

[0063] In this embodiment, by optimizing task scheduling according to the matching index, the tasks to be processed are assigned to the most suitable server to avoid resource waste or task delay. Especially when the scale of the virtual power plant is expanded, the load balance of each server can be guaranteed, and the processing capacity of the entire system can be improved.

[0064] In summary, the embodiment of the present application transfers tasks according to the compatibility index between servers, so that the computing resources of each server are fully utilized, avoiding the situation where some servers are idle while the resources of other servers are exhausted.

[0065] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0066] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0067] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0068] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0069] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0070] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A layered virtual power plant transaction optimization scheduling method, characterized in that: The following steps are involved: The virtual power plant control center obtains the virtual power plant working data and pre-processes the virtual power plant working data; Analyze the transaction saturation index of the server corresponding to each virtual power plant based on the pre-processed virtual power plant work data, where the transaction saturation index is used to reflect the saturation degree of transaction task processing in the virtual power plant server; When the transaction saturation index is higher than the first saturation threshold in the database, analyzing the transaction task transferability index of the corresponding server of the virtual power plant, wherein the transaction task transferability index is used to reflect the probability that a transaction task can be transferred to other virtual power plant servers; When the transaction saturation index is not higher than the second saturation threshold in the database, analyzing the transaction task receivable index of the server corresponding to the virtual power plant, wherein the transaction task receivable index is used to reflect the probability that a transaction task can be received by the current virtual power plant server; When the transaction saturation index is not higher than the first saturation threshold and higher than the second saturation threshold, no processing is performed on the virtual power plant server corresponding to the transaction saturation index; Analyzing the compatibility index of the corresponding servers of the two virtual power plants according to the transaction task transferability index and the transaction task receivability index, wherein the compatibility index is used to reflect the probability that the corresponding servers of the two virtual power plants can interact with each other in transaction tasks; Adjust transaction tasks between virtual power plant servers based on the compatibility index.

2. The hierarchical virtual power plant transaction optimization scheduling method according to claim 1, characterized in that: The virtual power plant control center serves as the upper layer of the virtual power plant, and the virtual power plant control center connects and manages the servers corresponding to each virtual power plant; The server serves as the middle layer of the virtual power plant, and the server corresponds to the virtual power plant one by one; The virtual power plant as the lower layer of the virtual power plant includes various power equipment; The virtual power plant control center exchanges data with each virtual power plant server through a communication protocol; In the virtual power plant control center, the virtual power plant working data from the virtual power plant server is obtained and summarized; The virtual power plant control center conducts centralized scheduling based on the status and load conditions of different virtual power plant servers; Each virtual power plant server collects data on the power equipment it includes.

3. The hierarchical virtual power plant transaction optimization scheduling method according to claim 1, characterized in that: The specific method of obtaining the transaction saturation index is as follows: Based on the preprocessed virtual power plant working data, the task quantity, task complexity, response time, and task execution time of the virtual power plant server are extracted; Obtain the weights of the number of tasks, task complexity, response time, and task execution time on the transaction saturation index according to the objective weighting method; Based on the number of tasks, task complexity, response time, task execution time and their weights, the transaction saturation index is derived using the transaction saturation index formula.

4. The hierarchical virtual power plant transaction optimization scheduling method according to claim 3 is characterized in that: The transaction saturation index formula is: In the formula, SD i is the transaction saturation index of the i-th virtual power plant server, i is the number of the virtual power plant server, i = 1, 2, ..., I, I is the total number of virtual power plant servers, e is a natural constant, TC i is the number of tasks of the i-th virtual power plant server, ONN i is the task complexity of the current task of the i-th virtual power plant server, TR i is the response time of the i-th virtual power plant server, TE i is the task execution time of the i-th virtual power plant server, α1 is TC i For SD i The weight of ONN i For SD i The weight of TR i For SD i The weight of α4 is TE i For SD i The weight of .

5. The hierarchical virtual power plant transaction optimization scheduling method according to claim 1, characterized in that: The specific method of obtaining the transferable index of the transaction task is as follows: Based on the pre-processed virtual power plant working data, the current load of the virtual power plant server, the remaining capacity of the server, and the resource availability are extracted; Obtain the weights of current load, remaining server capacity, and resource availability on the transferability index of transaction tasks according to the objective weighting method; Based on the current load, remaining capacity of the server, resource availability and their weights, the transaction task transferability index is derived through the transaction task transferability index formula.

6. The hierarchical virtual power plant transaction optimization scheduling method according to claim 5, characterized in that: The transferable index formula of the transaction task is: RT i =sinh[β1*TP i +β2*(1-SC i )+β3*(1-CR i )] Where RT i is the transferable index of the transaction task of the i-th virtual power plant server, i is the number of the virtual power plant server, i = 1, 2, ..., I, I is the total number of virtual power plant servers, TP i is the current load of the i-th virtual power plant server, SC i is the remaining capacity of the server in the i-th virtual power plant, CR i is the resource availability of the i-th virtual power plant server, β1 is TP i For RT i The weight of , β2 is SC i For RT i The weight of β3 is CR i For RT i The weight of .

7. The hierarchical virtual power plant transaction optimization scheduling method according to claim 1, characterized in that: The specific method of obtaining the receivable index of the transaction task is as follows: Based on the preprocessed virtual power plant working data, the remaining capacity, current load, average task execution time, and complexity of tasks to be received of the virtual power plant server are extracted; According to the objective weighting method, the weights of the server's remaining capacity, current load, average task execution time, and complexity of tasks to be received on the transaction task receivable index are obtained; Based on the remaining capacity of the server, the current load, the average task execution time, the complexity of the tasks to be received and their weights, the transaction task receivable index is calculated using the transaction task receivable index formula.

8. The layered virtual power plant transaction optimization scheduling method according to claim 7 is characterized in that: The formula for the receivable index of the transaction task is: RR i =tanh -1 [δ1*(1-TP i )+δ2*SC i +δ3*(1-TA i )+δ4*(1-ON i )] In the formula, RR i is the transaction task receivable index of the i-th virtual power plant server, i is the number of the virtual power plant server, i = 1, 2, ..., I, I is the total number of virtual power plant servers, SC i is the remaining capacity of the server of the i-th virtual power plant server, TP i is the current load of the i-th virtual power plant server, TA i is the average execution time of tasks on the i-th virtual power plant server, ON i is the complexity of the tasks to be received by the i-th virtual power plant server, δ1 is TS i For RR i The weight of ,δ2 is SC i For RR i The weight of , δ3 is TA i For RR i The weight of δ4 is ON i For RR i The weight of .

9. The hierarchical virtual power plant transaction optimization scheduling method according to claim 1, characterized in that: The specific method of obtaining the fit index is as follows: Obtaining a transaction task transferability index, a transaction task receivability index, and server compatibility data, wherein the server compatibility data includes hardware configuration similarity and communication delay; Obtain the weights of the transaction task transferability index, transaction task receivability index, and server compatibility data for the fit index through an objective weighting method; Based on the transaction task transferability index, transaction task receivability index, server compatibility data and their corresponding weights, the compatibility index is calculated through the compatibility index formula; The fit index formula is: In the formula, For the i a and the i-th b The matching index of the virtual power plant server, i is the number of the virtual power plant server, i = 1, 2, ..., i a ,...,i b ,...,I, I is the total number of virtual power plant servers, i a is the number of the virtual power plant server to which the task is to be transferred. b is the number of the virtual power plant server to receive the task. For the i a The transaction task transferability index of the virtual power plant server, For the i a The transaction tasks of the virtual power plant server can receive index, For the i a and the i-th b The hardware configuration similarity between the virtual power plant servers, For the i a The communication delay of the virtual power plant servers, For the i b The communication delay of the virtual power plant server, μ1 is for The weight of μ2 is for The weight of μ3 is the server compatible data for The weight of .

10. The layered virtual power plant transaction optimization scheduling method according to claim 1, characterized in that: The specific steps of adjusting the transaction tasks between the virtual power plant servers according to the matching index are: When the matching index exceeds the matching threshold in the database, the two groups of servers with the highest matching index are selected to transfer the tasks to be received to achieve optimized scheduling; When the pending task is received, the real-time status of the server is updated.

Images