Application of Fuzzy-NSGA-II for achieving maximum biodiesel yield from waste cooking oil.

Abstract

The increasing demand for renewable energy and efficient waste management has highlighted the need for innovative biodiesel production techniques. This study optimises biodiesel production from waste cooking oil (WCO) using fuzzy modelling and non-dominated sorting genetic algorithm-II (NSGA-II). The optimisation process focuses on key input parameters: methanol quantity, reaction temperature, reaction time, and catalyst concentration, which were normalised and represented using linguistic variables. Fuzzy logic was employed to predict biodiesel yield, expressed in terms of linguistic variables, and defuzzified to yield crisp output values. The developed model achieved a high R² value of 96.34%, demonstrating a strong correlation between input variables and biodiesel yield. The NSGA-II algorithm was utilised for multi-objective optimisation, determining the optimal conditions for biodiesel production: 150 ml of methanol, a reaction temperature of 62 °C, a reaction time of 63 min, and a catalyst concentration of 7.5 g. These parameters resulted in a maximum biodiesel yield of 97.36%. The Box-Behnken experimental design validated the model's efficiency, achieving a yield of 96.88%. This study emphasises the practical implications of optimised biodiesel production, such as reducing environmental pollution by recycling WCO and minimising reliance on fossil fuels. The optimised process meets ASTM standards and exhibits scalability potential for industrial-level production with minor modifications. The model's robustness makes it suitable for integration into intelligent manufacturing systems, ensuring consistent biodiesel quality and yield through automated monitoring and control mechanisms. Despite its success, challenges such as feedstock variability and initial setup costs must be addressed. Future studies should focus on adaptive models and energy-efficient processing technologies to enhance scalability and sustainability. This research demonstrates a significant step towards sustainable biofuel production, combining waste management with renewable energy generation.