ELM Filtering Algorithm Using Only Langmuir Probe Signals

A novel edge localized mode (ELM) filtering algorithm for fixed Langmuir probes (LP) data has

been developed in Python that automates the process without relying on external signals. The

filter employs adaptive peak height and width detection to ensure that varying types of ELMs

within a plasma shot can be captured. This technique improves the probes’ steady-state inter-

ELM data accuracy and can be used when other fast ELM detection signals (e.g., D-alpha filterscopes)

are unavailable. The high confinement mode (H-mode) in tokamaks is prone to ELMs,

uncontrolled periodic bursts of energy and particles. These energy and particle fluxes contribute

to undesirable plasma-material interactions. While existing tokamaks can largely withstand

ELMs, future devices will experience more intense ELMs which could cause significant damage

in much shorter time than the desired operating lifetime. DIII-D employs LPs in the divertors to

monitor edge-plasma conditions; ELMs appear as spikes in the LP signals, which are then

propagated to the derived quantities (n e , T e , V f , J sat ). These spikes can be orders of magnitude

greater than the inter-ELM values, skewing the probes’ time-averaged data. Results of ELM-

filtered LP data from the algorithm are compared to existing ELM filtering methods.