CN-122026335-A - Power distribution network control method and system based on bionic group intelligence

CN122026335ACN 122026335 ACN122026335 ACN 122026335ACN-122026335-A

Abstract

The invention discloses a power distribution network control method and system based on bionic group intelligence, and belongs to the field of power system control, wherein the method comprises the steps of constructing intelligent bodies for distributed resources in a power distribution network, exchanging information vectors with neighbors based on real-time state vectors thereof, and generating a cooperatively-oriented second neighbor exchanging information vector by combining a local potential field function; and finally, weighting and fusing the vector and the vector to form a control instruction, and realizing collaborative optimization control on the distributed resources. Therefore, by implementing the invention, the problem that the control of the power distribution network is difficult to ensure the control effect and improve the control robustness in the prior art can be solved.

Inventors

CHEN YUEJUN
LIU FANGZHOU
JI QINGFENG
CHENG XIANG
XIA TONG
JI AOYING
JIANG CHANG
YE JICHAO
ZHANG HANBING
HU XINWEI
LIU LINPING
LIU BIN
XU YONGHAI
HUANG HUI

Assignees

国网浙江省电力有限公司丽水供电公司

Dates

Publication Date: 20260512
Application Date: 20260413

Claims (10)

1. A power distribution network control method based on bionic group intelligence is characterized by comprising the following steps: Acquiring real-time data information corresponding to each distributed resource in the power distribution network, and constructing intelligent agents corresponding to each distributed resource, first neighbor exchange information vectors corresponding to each intelligent agent and real-time state vectors corresponding to each intelligent agent according to each distributed resource and the real-time data information corresponding to each distributed resource in the power distribution network; According to the real-time state vector and the first neighbor exchange information vector corresponding to each intelligent agent, a local potential field function construction method is adopted to obtain a second neighbor exchange information vector; Traversing each agent, if the second neighbor exchanging information vector corresponding to the currently traversed agent is an empty set, generating an autonomous optimization quantity according to the real-time state vector corresponding to the currently traversed agent, otherwise, generating a virtual gravitation according to the second neighbor exchanging information vector corresponding to the currently traversed agent; And carrying out weighted fusion on the autonomous optimization quantity and the virtual gravitation corresponding to each intelligent agent to obtain each control instruction, and controlling each distributed resource in the power distribution network according to each control instruction.
2. The method for controlling a power distribution network based on bionic group intelligence according to claim 1, wherein the method for constructing a local potential field function is adopted to obtain a second neighbor exchange information vector according to the real-time state vector and the first neighbor exchange information vector corresponding to each intelligent agent, specifically comprises the following steps: Obtaining a local potential field function value corresponding to each intelligent agent according to the real-time state vector corresponding to each intelligent agent; And obtaining a second neighbor exchange information vector corresponding to each intelligent body according to each local potential field function value and each first neighbor exchange information vector, wherein the local potential field function value corresponding to any intelligent body in each intelligent body is obtained according to the voltage stabilizing potential field component, the frequency adjusting potential field component, the economic running potential field component, the adjusting capability potential field component and the constraint punishment item corresponding to the intelligent body.
3. The method for controlling a power distribution network based on bionic group intelligence according to claim 2, wherein the obtaining the second neighbor exchange information vector corresponding to each agent according to each local potential field function value and each first neighbor exchange information vector specifically comprises: and traversing each agent, if the first neighbor exchange information vector corresponding to the currently traversed agent is not null, combining the local potential field function value corresponding to the currently traversed agent and the first neighbor exchange information vector corresponding to the currently traversed agent to obtain a second neighbor exchange information vector corresponding to the currently traversed agent, otherwise, outputting the second neighbor exchange information vector corresponding to the currently traversed agent as a null set.
4. The power distribution network control method based on bionic group intelligence according to claim 1, wherein the generating an autonomous optimization quantity according to the real-time state vector corresponding to the currently traversed agent is specifically as follows: Calculating by adopting a gradient calculation formula according to the real-time state vector corresponding to the currently traversed intelligent agent to obtain a potential field gradient corresponding to the currently traversed intelligent agent; And generating the autonomous optimization quantity according to the potential field gradient corresponding to the currently traversed agent.
5. The power distribution network control method based on bionic group intelligence according to claim 1, wherein the virtual attractive force is generated according to the second neighbor exchange information vector corresponding to the currently traversed agent, specifically: determining each neighbor agent according to the second neighbor exchange information vector corresponding to the currently traversed agent; And determining each virtual gravitation component according to the second neighbor exchange information vector corresponding to the currently traversed agent and the second neighbor exchange information vector corresponding to each neighbor agent, and carrying out vector synthesis on each virtual gravitation component to obtain the virtual gravitation.
6. The power distribution network control system based on bionic group intelligence is characterized by comprising a data acquisition module, a first processing module, a second processing module and a control module; The data acquisition module is used for acquiring real-time data information corresponding to each distributed resource in the power distribution network, and constructing each intelligent agent corresponding to each distributed resource, a first neighbor exchange information vector corresponding to each intelligent agent and a real-time state vector corresponding to each intelligent agent according to each distributed resource in the power distribution network and the real-time data information corresponding to each distributed resource; The first processing module is used for obtaining a second neighbor exchange information vector by adopting a local potential field function construction method according to the real-time state vector corresponding to each intelligent agent and the first neighbor exchange information vector; The second processing module is used for traversing each agent, generating an autonomous optimization quantity according to the real-time state vector corresponding to the currently traversed agent if the second neighbor exchange information vector corresponding to the currently traversed agent is an empty set, and generating a virtual gravitation according to the second neighbor exchange information vector corresponding to the currently traversed agent if the second neighbor exchange information vector is not the empty set; The control module is used for carrying out weighted fusion on the autonomous optimization quantity and the virtual gravitation corresponding to each intelligent agent to obtain each control instruction, and controlling each distributed resource in the power distribution network according to each control instruction.
7. The bionic population intelligent-based power distribution network control system according to claim 6, wherein the first processing module comprises a first processing unit and a second processing unit; the first processing unit is used for obtaining a local potential field function value corresponding to each intelligent agent according to the real-time state vector corresponding to each intelligent agent; The second processing unit is configured to obtain a second neighbor exchange information vector corresponding to each agent according to each local potential field function value and each first neighbor exchange information vector, where the local potential field function value corresponding to any one agent in each agent is obtained according to the voltage stabilizing potential field component, the frequency adjusting potential field component, the economic operation potential field component, the adjusting capability potential field component and the constraint penalty term corresponding to the agent.
8. The power distribution network control system based on bionic group intelligence according to claim 7, wherein the second neighbor exchange information vector corresponding to each agent is obtained according to each local potential field function value and each first neighbor exchange information vector, specifically: and traversing each agent, if the first neighbor exchange information vector corresponding to the currently traversed agent is not null, combining the local potential field function value corresponding to the currently traversed agent and the first neighbor exchange information vector corresponding to the currently traversed agent to obtain a second neighbor exchange information vector corresponding to the currently traversed agent, otherwise, outputting the second neighbor exchange information vector corresponding to the currently traversed agent as a null set.
9. The biomimetic population intelligent based power distribution network control system of claim 6, wherein the second processing module comprises a first processing subunit and a second processing subunit; The first processing subunit is used for calculating by adopting a gradient calculation formula according to the real-time state vector corresponding to the currently traversed intelligent agent to obtain a potential field gradient corresponding to the currently traversed intelligent agent; the second processing subunit is configured to generate the autonomous optimization quantity according to a potential field gradient corresponding to the currently traversed agent.
10. The biomimetic population intelligent based power distribution network control system of claim 6, wherein the second processing module further comprises a third processing subunit and a fourth processing subunit; the third processing subunit is used for determining each neighbor agent according to the second neighbor exchange information vector corresponding to the currently traversed agent; The fourth processing subunit is configured to determine each virtual gravity component according to the second neighbor exchange information vector corresponding to the currently traversed agent and the second neighbor exchange information vector corresponding to each neighbor agent, and perform vector synthesis on each virtual gravity component to obtain the virtual gravity.

Description

Power distribution network control method and system based on bionic group intelligence Technical Field The invention relates to the field of power systems, in particular to a power distribution network control method and system based on bionic group intelligence. Background In a modern power system, a power distribution network is used as a key hub for connecting a power transmission network and a user side, and is used for bearing the access and coordination tasks of multi-form resources such as a distributed power supply, an energy storage device and a flexible load, and the power distribution network control is realized by comprehensively controlling the output and adjustment of various resources, so that multi-objective optimization such as voltage stability, frequency adjustment, economic operation, adjustment capacity maintenance and the like is realized, and the reliability, flexibility and high efficiency of power supply are ensured. In the prior art, the group control of the distribution network mainly adopts two modes of centralized control or traditional complete distributed control, because the centralized control relies on global information acquisition and centralized decision, the bandwidth requirement of a communication link is extremely high and the single point failure risk of a central node exists, and the traditional complete distributed control usually needs high-frequency and full-quantity information exchange among intelligent agents, so that the communication cost is high, the problems that the system is difficult to adapt to the strong time variability and uncertainty caused by high-proportion distributed energy access, the popularization is limited in a large-scale scene, the robustness is insufficient when the abnormal conditions such as the limitation of the communication link and the node failure are faced, the group control failure is easy to be caused, and the local voltage and frequency fluctuation of a power grid and even the large-scale operation risk are caused. Disclosure of Invention The invention provides a power distribution network control method and system based on bionic group intelligence, which can solve the problem that in the prior art, the control effect is difficult to ensure and the control robustness is improved. In a first aspect, an embodiment of the present invention provides a power distribution network control method based on bionic group intelligence, including: Acquiring real-time data information corresponding to each distributed resource in the power distribution network, and constructing intelligent agents corresponding to each distributed resource, first neighbor exchange information vectors corresponding to each intelligent agent and real-time state vectors corresponding to each intelligent agent according to each distributed resource and the real-time data information corresponding to each distributed resource in the power distribution network; According to the real-time state vector and the first neighbor exchange information vector corresponding to each intelligent agent, a local potential field function construction method is adopted to obtain a second neighbor exchange information vector; Traversing each agent, if the second neighbor exchanging information vector corresponding to the currently traversed agent is an empty set, generating an autonomous optimization quantity according to the real-time state vector corresponding to the currently traversed agent, otherwise, generating a virtual gravitation according to the second neighbor exchanging information vector corresponding to the currently traversed agent; And carrying out weighted fusion on the autonomous optimization quantity and the virtual gravitation corresponding to each intelligent agent to obtain each control instruction, and controlling each distributed resource in the power distribution network according to each control instruction. The method and the system realize representation and information interaction of the running state of the distributed resources by constructing the exclusive intelligent agent for each distributed resource and defining the targeted first neighbor exchange information vector and the real-time state vector, reduce communication burden caused by redundant data transmission, integrate the multi-objective optimization requirement of the power distribution network into the second neighbor exchange information vector by means of a local potential field function, enable the information interaction among the intelligent agents to have cooperative guidance, provide an information basis for group cooperation, and then ensure the autonomous decision-making capability of the intelligent agent when no neighbor interaction exists and realize group cooperative linkage when neighbor support exists by generating autonomous optimization quantity or virtual attraction, greatly improve the robustness and fault tolerance of the system under complex working conditions, an