AI based failure analysis for one type of switch gears in power systems
This paper aims to explore an AI-based failure analysis method for a critical type of switchgear in power systems. Switchgears play a vital role in power systems, and their failure can significantly impact the stable operation of the entire system. Therefore, accurate and rapid switchgear fail...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis-Master by Coursework |
| Language: | English |
| Published: |
Nanyang Technological University
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/168051 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | This paper aims to explore an AI-based failure analysis method for a critical type of
switchgear in power systems. Switchgears play a vital role in power systems, and
their failure can significantly impact the stable operation of the entire system.
Therefore, accurate and rapid switchgear failure analysis is essential for ensuring the
reliability and safety of power systems.
In this project, the method of combining BP neural network with deep reliability
network is used to model and model the working condition and failure of power
system. As a basic feedforward neural network, BP neural network has the advantages
of simple training, high prediction accuracy and strong practicability. DBN is a kind
of deep network based on finite Boltzmann machine, which has powerful performance
of expression and promotion. This method can deeply learn and model
high-dimensional data and improve the accuracy and efficiency of error diagnosis.
By comparing the experimental results of the two methods, we found that the DBN
exhibits a distinct advantage in diagnostic accuracy and generalization capabilities,
enabling a more precise diagnosis of switchgear failure types and locations.
Furthermore, we discussed the model's interpretability and real-time applicability to
ensure the reliability and practicality of failure analysis.
The research findings presented in this paper hold reference value and practicality for
failure analysis and maintenance in power systems. Our study applies AI techniques
to the failure analysis of a critical switchgear in power systems, improving the
analysis accuracy and efficiency for such equipment. Consequently, it provides a
certain level of assurance for the safe and stable operation of power systems. |
|---|