Modeling and optimization of performance and emissions of a diesel engine fueled with water-diesel emulsions containing metal-organic nanoparticles by machine learning

The present study aimed to model and optimize the performance and emission characteristics of a diesel engine fueled with water-diesel emulsions containing metal-organic framework nanoparticles using a combination of adaptive neural-fuzzy inference system with optimal algorithm particle swarm generation (PSO-ANFIS). The multi-purpose particle swarm algorithm (MOPSO) was used to optimize engine performance and fuel composition. Water inclusion rate, engine load, and metal-organic framework nanoparticle concentration were considered as input parameters of the model. Brake specific fuel consumption, brake thermal efficiency, CO, CO2, UHC, NOx, and smoke were considered as model outputs. Sixteen experimental data were used in modeling and optimization processes. The results showed that the developed PSO-ANFIS models could accurately predict the objective functions. There was a good agreement between all the target data and the output of the developed models. According to the optimization results, water-diesel emulsion fuel containing 26.27 ppm metal-organic framework nanoparticles and 4.14 wt% water under engine load 60.15% of the full-load operating level was found to be optimal conditions.
.