Abstract:Machine learning (ML) and artificial intelligence have proven to be an invaluable tool in tackling a vast array of scientific, engineering, and societal problems. The main drivers behind the recent proliferation of ML in practically all aspects of science and technology can be attributed to: (a) improved data acquisition and inexpensive data storage; (b) exponential growth in computing power; and (c) availability of open-source software and resources that have made the use of state-of-the-art ML algorithms widely accessible. The impact of ML on the field of low-temperature plasmas (LTPs) could be particularly significant in the emerging applications that involve plasma treatment of complex interfaces in areas ranging from the manufacture of microelectronics and processing of quantum materials, to the LTP-driven electrification of the chemical industry, and to medicine and biotechnology. This is primarily due to the complex and poorly-understood nature of the plasma-surface interactions in these applications that pose unique challenges to the modeling, diagnostics, and predictive control of LTPs. As the use of ML is becoming more prevalent, it is increasingly paramount for the LTP community to be able to critically analyze and assess the concepts and techniques behind data-driven approaches. To this end, the goal of this paper is to provide a tutorial overview of some of the widely-used ML methods that can be useful, amongst others, for discovering and correlating patterns in the data that may be otherwise impractical to decipher by human intuition alone, for learning multivariable nonlinear data-driven prediction models that are capable of describing the complex behavior of plasma interacting with interfaces, and for guiding the design of experiments to explore the parameter space of plasma-assisted processes in a systematic and resource-efficient manner. We illustrate the utility of various supervised, unsupervised and active learning methods using LTP datasets consisting of commonly-available, information-rich measurements (e.g. optical emission spectra, current–voltage characteristics, scanning electron microscope images, infrared surface temperature measurements, Fourier transform infrared spectra). All the ML demonstrations presented in this paper are carried out using open-source software; the datasets and codes are made publicly available. The FAIR guiding principles for scientific data management and stewardship can accelerate the adoption and development of ML in the LTP community.

A neural network for the analysis of Langmuir-probe characteristics

A Long Short-Term Memory Network for Plasma Diagnosis from Langmuir Probe Data

A Bidirectional Long Short-Term Memory Network for Electron Density Diagnostic with Double Probe

Using convolutional neural networks to detect edge localized modes in DIII-D from Doppler backscattering measurements

Interpreting neural networks trained to predict plasma temperature from optical emission spectra

Advancements in neural network techniques for electric and magnetic field reconstruction: Application to ion radiography

Neural networks for turbulent transport prediction in a simplified model of tokamak plasmas

Low-temperature plasma simulation based on physics-informed neural networks: frameworks and preliminary applications

Neural network approximated Bayesian inference of edge electron density profiles at JET

Improved analysis techniques for cylindrical and spherical double probes

Foundations of machine learning for low-temperature plasmas: methods and case studies

Classification of the L-, H-mode, and plasma-free state: Convolutional neural networks and variational autoencoders on the edge reflectometer for KSTAR

Neural Networks Analysis of Thermal Characteristics on Plate-Fin Heat Exchangers with Limited Experimental Data

Bayesian analysis of triple Langmuir probe measurements for the characterization of Hall thruster plasmas

Linear and nonlinear flame response prediction of turbulent flames using neural network models

A novel plasma optical emission spectroscopy method with neural network model for improving the accuracy of plasma diagnostics in low-temperature xenon plasma

Langmuir Probe Diagnosis of a Cold Atmospheric Plasma Jet

Neural network based model of radio-frequency hollow cathode discharge characterized using Particle-in-Cell/Monte Carlo collision simulation

Hybridizing Physics and Neural ODEs for Predicting Plasma Inductance Dynamics in Tokamak Fusion Reactors