Development and optimisation of Fe ₃ O ₄ /PEG magnetic nanomaterials for nano-biocatalysis systems using the MSNN–BOA approach

Growing in popularity as a substitute for products derived from petrochemicals, sustainable biofuel production helps to mitigate common issues like climate change and global warming brought on by excessive use of fossil fuels. This study proposes a novel hybrid computational methodology for designing and optimising Fe3O4/polyethylene glycol (PEG) magnetic nanoparticles in nano-biocatalysis systems. The proposed hybrid technique, termed as the mixed style neural network (MSNN)-Botox optimisation algorithm (BOA) approach, combines MSNNs with the BOA. The primary objective of this work is to investigate the catalytic performance of Fe3O4/PEG magnetic nanoparticles to increase catalytic efficiency, stability, and reusability. The MSNN model predicts the optimal reaction parameters for the catalytic process and its accuracy is improved by BOA which effectively tunes the neural network weight. Simulation in MATLAB shows good catalytic stability and sustained activity with a biodiesel yield of 92% at a Fe3O4 mass ratio of 0.6 and a high yield of 77.3% for reusability after five cycles. Moreover, the proposed MSNN–BOA model achieved a minimum mean squared error (MSE) of 0.09, outperforming conventional methods like artificial neural network (ANN), feed-forward neural network (FNN) and genetic algorithm-back propagation (GA-BP) neural network. Therefore, the results demonstrate that MSNN–BOA is a strong and effective hybrid approach for magnetic nanomaterial optimisation, allowing for accurate catalytic performance prediction and advancing sustainable nanobiocatalytic systems.