Evaluation of a diesel engine optimized by non-evolutionary NLPQL and evolutionary genetic algorithms and assessing second law efficiency: Analysis in exergy loss and chemical exergy

Abstract

This study concerns about exergy analysis for a diesel engine which has been undergone an optimization either NLPQL (Non-Linear Programming by Quadratic Lagrangian) or Genetic algorithms. This optimization process has increased IMEP (indicated mean effective pressure) and decreased ISFC (indicated specific fuel consumption), NO and SMD (Sauter mean diameter). This SMD decreasing has led to better air/fuel mixing and though better combustion which has caused more temperature inside the cylinder and it is clear that entropy is in direct relation to temperature. All the optimization process has been done in the AVL Fire software. By exporting the results from AVL Fire to EES software, it would be possible to analyze the exergy change during the combustion and investigate the effect of optimization on the second law efficiency. The results depict that in the best cases of NLPQL and Genetic algorithms, the entropy generation has decreased by 7.5% and 6.4%, respectively. By the way, by increasing IMEP and thus the indicated power, the second law efficiency has increased from 29.9% in the baseline model to 34.5% in the best Genetic algorithm and 35.5% in the best NLPQL case.