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

DQN (deep Q network)-based selection method for ultrasonic bird-repelling optimal frequency

An ultrasonic and frequency technology, applied in the field of selecting the optimal frequency for ultrasonic bird repelling, can solve the problems of ineffective bird repelling frequency, single bird repelling frequency, and inability to repel the affected birds in a targeted manner, so as to avoid falling into the local optimum frequency. figure of merit, speeding up the training process, avoiding the effect of searching some invalid spaces

Active Publication Date: 2020-11-24
GUANGDONG POWER GRID CO LTD +1
View PDF9 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a DQN-based method to overcome the above-mentioned existing problems of single frequency of bird repelling, randomly occurring invalid frequency of bird repelling, adaptation of bird repelling frequency by birds, etc., and failure to drive targeted birds. Selection method of optimal frequency for ultrasonic bird repelling

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
  • DQN (deep Q network)-based selection method for ultrasonic bird-repelling optimal frequency

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0038] Such as figure 1 For the embodiment of the selection method of a kind of DQN-based ultrasonic repelling bird optimum frequency of the present invention, comprise the steps:

[0039] S1: Initialize the weight θ of the Q neural network of the bird protection device, the threshold of the number of iterations, prior knowledge, experience pool and number of iterations;

[0040] S2: Update the weight θ of the Q neural network;

[0041]S3: Use Doppler radar to detect and read the state of the bird within the detection range s t ;

[0042] S4: State s for the bird t Make a judgment, if the bird stays within the detection range, continue to step S5, if the bird does not stay within the detection range, repeat steps S3-S4;

[0043] S5: Judging whether there is prior knowledge of bird repelling in this area, including the more effective frequency of ultrasonic bird repelling in this area; if yes, combine the prior knowledge to select the frequency with the largest Q value; oth...

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 relates to the technical field of ultrasonic bird repelling, in particular to a DQN (deep Q network)-based selection method for ultrasonic bird-repelling optimal frequency. The method specifically comprises steps as follows: S1, initializing weight theta, a threshold of iteration times, prior knowledge, experience pool and the iteration times of Q network of a bird damage protectiondevice, wherein the prior knowledge is obtained by experience summary of bird damage protection in a certain area and comprises birds causing the most severe bird damage and the most sensitive frequency range of the birds in the area; S2, updating weight theta of the Q neural network; S3, adopting Doppler radar for detection, and reading state st of the birds in the detection range; and S4, judging the state st of the birds, continuing S5 if the birds stay in the detection range, and repeating steps S3 and S4 if the birds do not stay in the detection range. According to the DQN-based selectionmethod for the ultrasonic bird-repelling optimal frequency, the problems that existing bird-repelling frequency is single, invalid bird-repelling frequency is produced randomly, the birds adapt to the bird-repelling frequency and the like are solved, and the phenomenon that the birds causing damage cannot be repelled specifically is eliminated.

Description

technical field [0001] The invention relates to the technical field of ultrasonic bird repelling, in particular to a method for selecting the optimal frequency of ultrasonic bird repelling based on DQN. Background technique [0002] With the development of social production and people's living standards, power grid users have gradually increased their requirements for power supply quality and stability. In recent years, the transformation and upgrading of power equipment has greatly improved the quality of power supply, and the reliability and stability of power supply have been guaranteed. However, accidents of power outage and tripping also happen occasionally, which have a certain impact on life and production. Bird activity is one of the main causes of transmission line failures. [0003] Since the habits of the bird species involved in the transmission line are not considered, the frequency parameters of the bird repellent are designed in a generalized manner. Among ...

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
Patent Type & Authority Applications(China)
IPC IPC(8): A01M29/18G06N3/04G06F16/9537
CPCA01M29/18G06F16/9537G06N3/045
Inventor 陈益平张羽周庆东蓝伟松甘团杰刘天绍张家耀吴华标周宇尧莫钜槐张经纬
Owner GUANGDONG POWER GRID CO LTD
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