Hard X-Ray Emission Detection Using Deep Learning Analysis of the Radiated UHF Electromagnetic Signal From a Plasma Focus Discharge

Abstract

A method to determine the presence of hard X-ray emission processes from a dense plasma focus (205 J, 22 kV, 6.5 mbar H 2) using Ultra High Frequency (UHF) measurements and deep learning techniques is presented. Simultaneously, the electromagnetic UHF radiation emitted from the plasma focus was measured with a Vivaldi UHF antenna, while the hard X-ray emission was measured with a scintillator-photomultiplier system. A classification algorithm based on deep learning methods, using two-dimensional convolutional layers, was implemented to predict the hard X-ray signal standard deviation value using only the antenna signal measurement. Two independent datasets, consisting of 999 and 1761 data pairs each, were used in the analysis. Different realizations of the training/validation process using a deep learning model, obtained overall better results in comparison to other machine learning methods like kneighbors, decision trees, gradient boost, and random forest. The results of the deep learning algorithm, and even its comparison with other machine learning methods, indicate that a relationship between the electromagnetic UHF radiation and hard X-ray emission can be established, enabling the indirect detection of hard X-ray pulses only using the UHF antenna signal. This indirect detection presents the opportunity to have a simple and low-cost diagnostic, compared to the methods currently used to characterize the pulses of X-rays emitted from plasma focus discharges. INDEX TERMS Deep learning, Plasma focus, UHF antenna, X-ray pulse.

Inference of X-Ray Emission From a Plasma Focus Discharge: Comparison Between Characteristic Parameters and Neural Network Analyses

“…The signal from both SC-PMT were analysed in terms of the normalized standard deviation value R with the same procedure as in [34], i.e. signal standard deviation S given by equation 1 and using the value S Base to normalize it, equation 2.…”

“…to find a non linear function whose output is a predicted real value [39], although this latter alternative has more physical sense in this context [38]. Given the discussion above, the classification problem of the R values was considered in this work continuing the framework established in [34]. Fig.…”

“…To complement the information of the performance metrics, the probability of achieving the full identification by mere chance was included, denoted by G. This probability was calculated with a binomial distribution that takes into account the number of correct full identification predictions k, the total of the validation set (n = 192 discharges) and a probability of 0.25 chance to guess one of the four classes [34], see equation 5.…”

“…EM burst diagnostic has been used either inside or outside the chamber [29]- [32]. Recently, the EM burst measurement has been correlated with the inductive measurement [33] and with the X rays detection using a scintillator-photomultiplier (PMT) system [34]. From these two works, the Vivaldi antenna, originally designed to measure the fast EM emission of partial discharges [35], has shown promising results.…”

“…To complement the use of characteristic parameters that electrically describes the discharge for the correlations with other phenomena [24], [27], the use of machine learning algorithms, in particular the deep learning approach [36], has shown to be practical in the field of plasma diagnostics [37], [38]. Recently, it was used in plasma focus to infer the conditions for X rays emission based on only the signal from a Vivaldi antenna [34].…”