Authors: Leon Ofman (CUA/NASA GSFC), Scott A Boardsen (UMBC/NASA GSFC), Viacheslav M Sadykov (GSU), Lan K Jian (NASA GSFC), Parisa Mostafavi (JHU/APL), Jaye L Verniero (NASA GSFC), Davin Larson (SSL/UC Berkeley), Roberto Livi (SSL/UC Berkeley), Michael McManus (SSL/UC Berkeley), Ali Rahmati (SSL/UC Berkeley), Michael L Stevens (SSL/UC Berkeley)

Recent observations by the Parker Solar Probe Solar Probe Analyzer Ions (SPAN-I) and FIELDS instrument find abundant evidence of ion-scale waves and non-Maxwellian ion velocity distribution functions (VDFs) with super-Alfvénic beams and large temperature anisotropy. The kinetic instabilities impact the eventual collisionless dissipation of turbulent fluctuations. We use 2.5D and 3D hybrid models of proton-alpha SW plasma to study the evolution and associated ion kinetic instabilities of the VDFs for several cases of interest in the super- as well as sub-Alfvénic solar wind. The hybrid models show the growth and the nonlinear saturation of the kinetic instabilities, extending the linear Vlasov stability analysis. We model the energy exchanged between protons, alpha particles, and magnetic field fluctuations associated with kinetic waves due to the relaxation of the instabilities. The models provide the complete 3D velocity space structure of the proton and alpha particles VDFs at the various stages of the instability progression in the SW frame, not limited by the data constraint of the single point spacecraft frame and the instrumental line-of-sight limits brought on by the PSP heat shield. Due to the abundant PSP data, the analysis would benefit from the use of automated identification of ion kinetic instability periods. We investigate the application of Artificial Intelligence Machine Learning (AI/ML) methods to automate the detection of unstable VDFs and classification of the kinetic instabilities. For this purpose, we use the hybrid modeling results of ion kinetic instabilities for the construction of the ML-ready data set.