Transient Pulse Signal Recognition and Extraction in Particle Accelerator Devices

Particle accelerator devices belong to complex high-tech large scientific engineering equipment, and the electromagnetic environment is exceptionally complicated. Transient pulse signals negatively impact the operation of equipment in particle accelerator devices through conduction emission, cable c...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2025 4th Asia Conference on Algorithms, Computing and Machine Learning (CACML) s. 1 - 5
Hlavní autoři: Yin, YongChang, Ma, DongMei, Liu, Ye, Feng, AnHui
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 28.03.2025
Témata:
Data mining
Electromagnetic compatibility
Linear particle accelerator
Machine learning algorithms
Object recognition
particle accelerator device
signal recognition
Spectrogram
Switches
Testing
Transient analysis
transient pulse signal
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Particle accelerator devices belong to complex high-tech large scientific engineering equipment, and the electromagnetic environment is exceptionally complicated. Transient pulse signals negatively impact the operation of equipment in particle accelerator devices through conduction emission, cable crosstalk, near-field coupling, and other means. Therefore, identifying transient pulse signals in particle accelerator devices is an important topic for alleviating electromagnetic compatibility issues in particle accelerators. This paper systematically analyzes the causes and main effects of transient pulse signals generated by particle accelerator devices and proposes a transient pulse signal identification algorithm based on the threshold method. First, a testing system is constructed based on the IEC-61000-4-4 standard to obtain the pulse signal spectrum; second, a graphical pixel processing method is used to denoise the test spectrum; finally, based on the threshold identification algorithm, transient pulses are extracted, and the accuracy of different thresholds in signal identification is compared using the signal-to-noise ratio to define the optimal threshold, providing support for interference signal identification and spectrum management in particle accelerator devices.
DOI:10.1109/CACML64929.2025.11010968