Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Power system reactive power optimization method based on depth determination strategy gradient reinforcement learning

A power system and reinforcement learning technology, applied in reactive power compensation, reactive power adjustment/elimination/compensation, electrical components, etc., can solve problems such as deep reinforcement learning is rarely applied

Active Publication Date: 2019-12-03
HARBIN INST OF TECH +1
View PDF5 Cites 33 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Today, intelligent control using deep reinforcement learning has been applied in various fields, and has achieved great success, such as AlphaGo, ATARI Game, robot control, etc., but deep reinforcement learning is rarely used in the field of power system optimization

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 reactive power optimization method based on depth determination strategy gradient reinforcement learning
  • Power system reactive power optimization method based on depth determination strategy gradient reinforcement learning
  • Power system reactive power optimization method based on depth determination strategy gradient reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0059] A power system reactive power optimization method based on depth-determined policy gradient reinforcement learning, the steps are as follows:

[0060] Step 1: Use the physical quantities used in the reactive power optimization calculation of the power system to describe the variables in the deep reinforcement learning, so as to achieve the purpose of applying the deep reinforcement learning algorithm to the reactive power optimization of the power system:

[0061]

[0062] Among them, P loss is the active network loss; k transformer ratio; n l is the total branch number of the network; G k(i,j) is the conductance of branch i– j; U i , U j are the voltages of nodes i and j respectively; ω i , ω j are the phase angles of nodes i and j respectively; f is the reactive power optimization purpose of the power system;

[0063] Step 2: The wide-area measurement system of the power system obtains the power, phase, power angle, and voltage amplitude information of each n...

Embodiment 2

[0104] In this embodiment, a power system reactive power optimization algorithm based on deep deterministic policy gradient reinforcement learning is used to perform reactive power optimization calculations on the IEEE30 node test system. The power system simulation part uses the Matlab program to calculate the power system power flow; the algorithm part uses Python language programming and is compiled and passed on the Pycharm compiler software. At the same time, the tensorflow1.0 deep learning framework developed by Google and the CUDA9.0 computing framework of Nvidia are used, and the parallel computing engine of the GTX-1060 graphics card is used to make the entire optimization process have extremely fast computing speed.

[0105] (1) IEEE30 node standard test system

[0106] The system has four transformers and five generators. Four compensation points are selected to meet the requirements of reactive power optimization. The upper and lower limits of the node voltage are ...

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 provides a power system reactive power optimization method based on depth determination strategy gradient reinforcement learning. A deterministic depth gradient strategy algorithm is applied to the traditional power system reactive power optimization problem. The voltage state of the power system is sensed through a depth neural network, a corresponding action decision is then made by using a reinforcement learning method, a correct generator terminal voltage adjustment action, a node capacitor bank switching action and a transformer tapping point adjustment action are made to adjust reactive power distribution in the power system, the active power network loss of the power system is minimized. As the neural network is divided into an online network and a target network, association between parameter updating and adjacent training in each training process of the neural network is avoided, and the problem that reactive power optimization of the power system is caught in local optimization is avoided. On the premise of conforming to the security constraint of the power system, the economical efficiency of the operation of the power system is improved by reducing the network loss of the power system.

Description

technical field [0001] The invention relates to the field of power system reactive power optimization, in particular to a power system reactive power optimization method based on depth determination strategy gradient reinforcement learning. Background technique [0002] The reactive power distribution of the power system will affect the power loss of the system and the voltage qualification rate of the nodes, so the reactive power optimization of the power system is an important means to improve the economic operation of the power system. Power system reactive power optimization is achieved by adjusting control settings in the grid, such as synchronous generator terminal voltage, node capacitor bank switching, transformer tap settings, etc. Power system reactive power optimization can be expressed as a mathematical optimization model. From an economic point of view, the goal of optimization is to minimize the network loss of the system. [0003] Power system reactive power ...

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): H02J3/18
CPCH02J3/1871Y02E40/30
Inventor 张伟杨丰毓钱敏慧陈宁赵大伟
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com