Abstract In aviation studies, optimizing jet engine performance is becoming increasingly prominent through experimental analysis. The importance of this study lies in obtaining experimental data from a micro turbojet engine by measuring several engine sensors. The main goal is to optimize engine performance variables by applying multi-objective algorithms. Based on experimental data, specific fuel consumption (SFC), thermal efficiency, exergy efficiency, and exergetic improvement rate (EIR) are modeled by employing artificial neural networks. The output parameters trained with four input variables (thrust, fuel consumption, pressure ratio, and exhaust gas temperature) are subjected to multi-objective genetic, multi-objective particle swarm, and multi-objective grey wolf algorithms. The obtained results are compared with each other. From idle speed to maximum speed, SFC varies between 0.466 kg Nh −1 and 0.178 kg Nh −1 , while thermal efficiency varies between 1.128% and 8.3738%. All four parameters are modeled with ANN with R 2 above 0.99. On the other hand, according to the optimization results, minimum SFC and maximum exergy efficiency are calculated as 0.1785 kg Nh −1 and 7.8452% with MOGA, whereas these are computed as 0.1783 kg Nh −1 and 7.6937% with MOPSO. The best result is obtained with MOGWO. Namely, thanks to MOGWO, SFC is found to be 0.1782 kg Nh −1 , whereas thermal efficiency is obtained as 8.0261%. The optimum input values providing the best values are observed as 92 N for thrust, 3.4829 for compressor pressure ratio, 559.8986 °C for EGT, and 0.0023 kg s −1 for fuel flow.
Kılıç et al. (2026) studied this question.