Advertisement
Published in:
2021 | OriginalPaper | Chapter
Traction Inverter Fault Detection Method Based on Welch and K-Nearest Neighbor Algorithm
Authors : Sara Zerdani, Mohamed Larbi El Hafyani, Smail Zouggar
Published in: Proceedings of the 2nd International Conference on Electronic Engineering and Renewable Energy Systems
Publisher: Springer Singapore
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Abstract
Most commonly, the existing fault diagnosis approaches depend on the availability of the measurements, and therefore, on the reliability of the sensor, consequently, if a fault occurs at the sensor level, this may result in a false alarm, indicating the occurrence of a failure in the energy conversion devices. The greatest extensive fault diagnosis techniques could estimate the defects and even located them, but they neglected the impact of the quality factor of the input instructions.
Regarding these outcomes, this paper suggests a new multi faults diagnosis algorithm based on the Welch method and the K-Nearest Neighbor classifier algorithm. In this approach, the Welch method is applied to estimate the power spectral density; it provides the foremost signal components that discriminate the deficiencies of the devices, and then the character of the shortages is identified using the K-Nearest Neighbor classifier, which is proper for multi-class labeling.
The effectiveness of the recommended strategy is confirmed via simulation, within its employment in the diagnosis of electric vehicle powertrain defects, indistinct, the traction inverter at the faulty and healthy status of the current sensor.