RESUMO
This study investigates the impact of adding diethyl ether (DEE) to pyrolysis oil derived from mixed plastic waste on engine performance, combustion characteristics, and emissions. The blending of different DEE concentrations (5%, 10%, and 15% by volume) with waste plastic oil (WPO) was analyzed. Experiments were conducted on a four-cylinder diesel engine, varying engine loads while maintaining engine speed. The results indicate that WPO mainly comprises middle-distillate hydrocarbons (52.58% C13-C18 and 26.15% C19-C23). While WPO had lower specific gravity, density, and flash point, it met diesel fuel specifications for kinematic viscosity and cetane index. The addition of DEE led to decreased properties in all blended fuels, except for the cetane index. Engine performance declined with WPO-DEE blends at low engine loads but improved at high engine loads with minimal variation as DEE concentration increased. DEE addition resulted in a shorter ignition delay and earlier combustion, although increasing DEE concentration did not further advance combustion. NOx emissions significantly decreased with DEE addition, while HC and CO emissions remained unaffected at high engine loads. To optimize the process, the non-dominated sorting genetic algorithm II (NSGA-II) with generalized regression neural networks (GRNNs) was employed as a surrogate multi-objective function. The GRNNs model demonstrated excellent performance, achieving high R2 values of 0.952 and 0.918, low RMSE values of 0.659 and 0.310, and MdAPE values of 2.675% and 5.098% for brake thermal efficiency (BTE) and NOx, respectively. The NSGA-II algorithm with GRNNs model proved successful in predicting the multi-objective function in the optimization process, even with limited data. The Pareto frontier analysis revealed an optimal DEE percentage of approximately 10% to 14% for maximum BTE and minimum NOx, with engine loads distributed around 30, 40, and 100 N m.
RESUMO
Today, artificial intelligence plays a huge role in the mechanical engineering field for solving many complex problems and the problem with fracture mechanics is one of them. In fracture mechanics, artificial intelligence is used to predict crack behavior under various conditions such as mixed-mode loading. Many parameters are used for explaining the crack behavior under various conditions, but those parameters are obtained from destructive testing, in which usually, only one data point is obtained from each test. An artificial problem method requires a large amount of data to train the model to be able to learn crack behavior, which is a disadvantage of applying this method to fracture mechanics. To eliminate the disadvantage of the large amount of experiment data required for modeling, in this study, the small data obtained from the experiment along with data obtained from fracture criteria that were used for elementary prediction of mixed mode fracture toughness were used to create an artificial intelligence model. Data from the experiment was combined with fracture criteria data using the multi-fidelity surrogate model that is described in this study. The mixed mode I/II fracture toughness of the PMMA material was tested in order to primarily propose the data combination technique. After the modeling process, the prediction results indicated that the performance of a model in which the actual test data was combined with the data from the fracture criteria (multi-fidelity surrogate model) was more predictively effective compared to only actual data-based modeling.
