Optimization of engine performance, emission and combustion parameters by using mixed nonedible oil biodiesel with nano-additives using hybrid techniques

Abstract

The current study is focused on three nonedible oils resource such as Ceiba pentandra, Mahua longifolia, and Azadirachta indica mixed together in a specific proportion for biodiesel production. The produced biodiesel serves as an alternative fuel source for operating a variable compression ratio diesel engine, addressing escalating environmental pollution and stringent emission standards. A nano-additive comprising CaO–TiO2 is incorporated into biodiesel blends to enhance engine performance and mitigate emissions. The matrix undergoes experimentation and validation using Box-Behnken Design, Deep Belief Network, and a hybrid DBN-based Arithmetic Optimization Algorithm to optimize engine parameters for achieving higher brake thermal efficiency, lower emission rates, and reduced brake-specific fuel consumption. All the experiments are conducted by varying the engine load, speed, quantity of nano-additives and biodiesel blend. The final empirical model is more significant for DBN with a regression coefficient greater than 0.99. The converged optimized input parametric ranges are load 60 %, biodiesel 25 %, nano-additive 20 wt% and engine speed 1400 rpm. When comparing diesel fuel with CaO–TiO2 biodiesel, the proposed method resulted in significant improvements, including a 41 % reduction in brake-specific fuel consumption, a 32 % decrease in CO emissions, a 25 % decrease in NOX emissions, and a 16.9 % increase in brake thermal efficiency.