Metaheuristic Algorithms Based Comparative Study Integrated with Artificial Intelligence to Forecast Shelf Life (Serviceability) of Propellants

Abstract

The prediction of ageing and serviceability in single base propellants has become remarkably effortless owing to the rapid advancement of new analytical techniques, specifically High-performance liquid chromatography (HPLC). However, HPLC only provides momentary situations. To address this issue and to obtain real time ageing prediction of SBPs this research explores the use of machine learning (ML) integrated with optimization techniques like genetic algorithms (GA) and particle swarm optimization (PSO). Ageing refers to the deterioration of the propellant over time, which can affect its functionality and performance. The study uses predictive ML models to optimize, automate, and surveil single-base propellants in combination with GA and PSO to enhance their performance. Several criteria are used in this study to evaluate the models' accuracy and probability for prediction. The study is significant with 0.89 coefficient of determination for the optimum performing ML technique namely ET-GA to forecast the effects of ageing on propellants, contributing to the advancement of testing and surveillance techniques for single-base propellants. An optimized ML model ET-GA shows maximum of 5% deviation with experimentation is used to create a Graphical User Interface (GUI) that simplifies the calculation of the remaining effective stabilizer percentage (RES %).