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

CNN (Convolutional Neural Network)-based method for quickly identifying low-frequency oscillation modal characteristics of power system

A low-frequency oscillation and power system technology, applied in neural learning methods, character and pattern recognition, electrical components, etc., can solve problems such as insufficient consideration of the non-stationary characteristics of the measured oscillation signal

Active Publication Date: 2021-02-05
STATE GRID SICHUAN ECONOMIC RES INST
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these studies all have the defect of insufficient consideration of the non-stationary characteristics of the measured oscillation signal.

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
  • CNN (Convolutional Neural Network)-based method for quickly identifying low-frequency oscillation modal characteristics of power system
  • CNN (Convolutional Neural Network)-based method for quickly identifying low-frequency oscillation modal characteristics of power system
  • CNN (Convolutional Neural Network)-based method for quickly identifying low-frequency oscillation modal characteristics of power system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0119] In order to verify whether the algorithm of the present invention can identify the power system superimposing new oscillation modes in the oscillation process, the ideal LFO test signal is constructed as follows:

[0120]

[0121] In formula (9), ε(t) represents a step function, and η(t) represents a noise signal.

[0122] The LFO signal image as Figure 4 As shown, the signal length is 12s, and the signal-to-noise ratio SNR=10dB. Before 5s, the signal contains two modes, where the frequency f 1 = 0.84Hz, f 2 =1.21Hz, attenuation factor σ 1 =0.12, σ 2 =-0.15. When t=5s, a new oscillation mode is introduced, its frequency f 3 = 1.58Hz, attenuation factor σ 1 =0.3. After 6s, due to the σ in the initial mode 2 =-0.15 belongs to the strong attenuation component, and this mode is no longer counted as the dominant mode at this time, so the signal still contains two modes, where the frequency f 1 = 0.84Hz, f 3 = 1.58Hz, attenuation factor σ 1 =0.12, σ 2 = 0.3. ...

Embodiment 2

[0129] In order to verify the actual identification effect of the present invention, a piece of LFO measured data is obtained from the power system. Such as Figure 5 As shown in , this section of LFO signal is excited by two small disturbances, located at 4s and 7s respectively. In order to identify the LFO modal characteristics excited after the two disturbances, the data after the disturbance is intercepted as the LFO signal to be tested in this embodiment.

[0130] Step 7 acquires samples of the LFO signal to be tested through the sliding time window. The sliding window length is 4s, the sliding interval is 1s, and the sampling frequency is 100Hz. In order to verify the applicability of the present invention in the case of multiple disturbances in the system, it is necessary to use sliding window sampling after two disturbances occur. In this embodiment, two segments of signals are selected as identification objects, respectively 4.2-8.2s and 7.05-11.05s. Among them, 4...

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

PropertyMeasurementUnit
Snraaaaaaaaaa
Login to View More

Abstract

The invention discloses a CNN-based method for quickly identifying low-frequency oscillation modal characteristics of a power system, and provides a method which is more suitable for identifying relatively complex operating conditions and oscillation environments in a high-proportion renewable energy grid-connected power electronic system, and comprises the following steps of: generating LFO sample data according to an EDSs mathematical model; carrying out preprocessing operation on the LFO sample data by adopting a time domain feature extraction algorithm; determining a classification criterion according to identification requirements, and dividing and marking LFO sample data according to the classification criterion for subsequent network training and testing; building a CNN model, inputting a training sample for network training, and determining whether the network training is completed or not through testing the classification accuracy of the sample; inputting an LFO signal to be detected into the CNN through sliding window sampling, and completing identification of LFO frequency and attenuation factor modal characteristics through output analysis. The method has the advantagesthat the low-frequency oscillation modal characteristics of the power system can be quickly identified and the like.

Description

technical field [0001] The invention relates to the technical field of power system stability and control, in particular to a method for quickly identifying modal characteristics of low frequency oscillation (LFO) in a power system based on a convolutional neural network (CNN). Background technique [0002] With the increasing scale of the power grid, the interconnection of long-distance AC and DC systems, and the increase in the proportion of renewable energy grid-connected power generation, the problem of low-frequency oscillation in the power system has become increasingly prominent, seriously endangering the safe and stable operation of the power system. In recent years, the occurrence of low-frequency oscillations in power systems has emerged in an endless stream. Once low-frequency oscillation occurs, it will often trigger the distance protection of the transmission line, and in severe cases, it will damage the equipment and even cause the system to disconnect. Theref...

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): G06K9/00G06N3/04G06N3/08H02J3/00
CPCG06N3/08G06N3/084H02J3/002H02J2203/20G06N3/045G06F2218/08G06F2218/12Y02E60/00
Inventor 叶圣永张文涛魏俊刘旭娜刘立扬韩宇奇龙川刘洁颖李达赵达维
Owner STATE GRID SICHUAN ECONOMIC RES INST
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