Power system automatic operation method and device based on physical information and deep reinforcement learning

A technology of power system and physical information, applied in the field of automatic operation method and device of power system, can solve the problem that human operators cannot effectively monitor and manage real-time, etc., and achieve both economy and effectiveness, real-time stable control, and good effect. Effect

Pending Publication Date: 2021-12-07
STATE GRID JIANGSU ELECTRIC POWER CO ELECTRIC POWER RES INST +1
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

As grid complexity increases, human operators cann

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Power system automatic operation method and device based on physical information and deep reinforcement learning
  • Power system automatic operation method and device based on physical information and deep reinforcement learning
  • Power system automatic operation method and device based on physical information and deep reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0061] Construct test cases with IEEE14 node model data, and use power system status information data including 36 lists of each information type as a reinforcement learning environment, such as line current, power generation, power consumption, sudden weather disasters, etc. for running simulations . The goal of reinforcement learning optimization is to consider factors such as voltage crossing, current thermal threshold, and transmission line status, and try to avoid the power system from entering the end state.

[0062] The action space includes switching on or off specific transmission lines and reconfiguring substation connection topology. The complete action space has N A = 76 elements, but can be reduced to 56, a single action consists of integers A∈[0,N A ) flag, which maps to any configurable substation or switchable transmission line.

[0063] The neural network structure used to train the DQN agent is shown in Table 1, and the parameters of the test set and verif...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a power system automatic operation method and device based on physical information and deep reinforcement learning. The method comprises the following steps: setting a neural network structure, parameters, a state space and a selectable action space required by reinforcement learning training, constructing a deep Q learning intelligent agent, and predicting automatic actions by minimizing frequency and severity of line faults; directly using physical information of a power system as a known limit of training, reducing training data, and shortening training time; and designing a reward function for optimization, training to select different actions to achieve the maximum reward, and searching for the optimal automatic operation. According to the invention, characteristics of the physical information are combined, optimization control is carried out on power transmission lines and transformer substation topology of the power system based on a DQN algorithm, economy and effectiveness are both considered; and the stable control of the power system can be carried out in real time under the condition of less data.

Description

technical field [0001] The invention belongs to the technical field of power system operation optimization, and in particular relates to a power system automatic operation method and device based on physical information and deep reinforcement learning. Background technique [0002] With the increase of demand for electricity due to economic and social development, the massive access of wind energy and intermittent renewable energy has put the power grid under pressure on power supply. Power outages are highly destructive to the power system, catastrophic to the safe and stable operation of the power grid, and the cost of corrective maintenance is extremely high. As grid complexity increases, human operators cannot effectively monitor and manage real-time issues. Strong demand growth, limited generation capacity, and complex real-time operations provide favorable conditions for the use of intelligent algorithms to participate in power system decision-making. [0003] The re...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06F30/27G06F119/02
CPCG06F30/27G06F2119/02
Inventor 袁晓冬袁宇波孙天奎史明明司鑫尧苏伟郭佳豪姜云龙肖小龙
Owner STATE GRID JIANGSU ELECTRIC POWER CO ELECTRIC POWER RES INST
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap