Data Science and Machine Learning Processes for IoT-Based Pulsed Plasma Thruster Research

Abstract

Pulsed plasma thrusters (PPTs) have very high specific impulse but low efficient electric propulsion engine, which are being used as primary or secondary propulsion mechanism for spacecrafts (cube, micro, nano, or pico satellites). PPTs are used in academic and/or industrial research from decades, but its experimental data collection methods are normally offline, analysis methods are conventional, and data interpretation lies in electromagnetic, physical, and chemical domain. It is inadequate to explain PPT-generated data in few specific domains only, and it may prone to inadequate explanations, which can hide vast insights of data. Actually, PPT-generated data are usually big data. It is essential to analyze and explain PPT data in data science and machine learning (ML) domain to get keen insights from the data. To this date, no such complete solution exists for PPT either in industrial arena or academia to use all these three technologies. To meet the gap, we propose and implement an Internet of Things (IoT)-based architecture for PPT experimental facility to collect, accumulate, and process PPT data intelligently. This architecture yields some ML models with prediction accuracy above 93%. This architecture is a complete IoT and big data-based artificial intelligence implementation on PPT data through data science and ML processes. This architecture is capable to contribute in both academic and industrial research, development, and deployment.

Received: 9 August 2023 | Revised: 20 November 2023 | Accepted: 27 December 2023

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

Data available on request from the corresponding author upon reasonable request.