M‐NLP Inference Models Using Simulation and Regression Techniques. Issue 2 (27th January 2023)

• Record Type: Journal Article
• Authors: Liu, Guangdong; Marholm, Sigvald; Eklund, Anders J.; Clausen, Lasse; Marchand, Richard
• Is Part Of: Journal of geophysical research. Volume 128:Issue 2(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2023-01-27
• Subjects: multi‐needle Langmuir probe -- particle in cell simulation -- multivariate regression -- machine learning inference; Magnetospheric physics -- Periodicals; Space environment -- Periodicals; Cosmic physics -- Periodicals; Planets -- Atmospheres -- Periodicals; Heliosphere (Astrophysics) -- Periodicals; Geophysics -- Periodicals; 523.01
• DOI: 10.1029/2022JA030835
• Languages: English
• ISSNs: 2169-9380

Abstract

Abstract: Current inference techniques for processing multi‐needle Langmuir probe (m‐NLP) data are often based on adaptations of the Orbital Motion‐Limited (OML) theory which relies on several simplifying assumptions. Some of these assumptions, however, are typically not well satisfied in actual experimental conditions, thus leading to uncontrolled uncertainties in inferred plasma parameters. In order to remedy this difficulty, three‐dimensional kinetic particle in cell simulations are used to construct a synthetic data set, which is used to compare and assess different m‐NLP inference techniques. Using a synthetic data set, regression‐based models capable of inferring electron density and satellite potentials from 4‐tuples of currents collected with fixed‐bias needle probes similar to those on the NorSat‐1 satellite, are trained and validated. The regression techniques presented show promising results for plasma density inferences with RMS relative errors less than 20%, and satellite potential inferences with RMS errors less than 0.2 V for potentials ranging from −6 to −1 V. The new inference approaches presented are applied to NorSat‐1 data, and compared with existing state‐of‐the‐art inference techniques. Plain Language Summary: We present a new technique for processing measurements from a set of cylindrical Langmuir probes with different fixed voltages relative to a satellite in low Earth orbit. Our approach, based on results from 3‐D simulations, does not rely on the assumptions made in currently used theoretical models, and it accounts for more physical processes than possible in theories. Our goal is to improve the accuracy of plasma density and satellite potential inferences using this type of instrument, while providing uncertainties in the inferences. Our inference models are applied to NorSat‐1 satellite data and are compared with currently used theoretical model inferences. It is found that our model inference is more consistent with satellite data. Key Points: 3‐D kinetic PIC simulations are used to simulate currents collected by multi‐needle Langmuir probe (m‐NLP) in order to create a synthetic solution library Models to infer physical parameters from m‐NLP measurements are constructed and assessed on the basis of synthetic and in situ data sets Promising new approaches are identified to analyze m‐NLP measurements based on simulation and in situ data