The New Hybrid Meta-Model by Using DOE and RSM for Liquid Propellant Engine’s Feed System Optimization

The present paper strives for optimization of the Liquid-Propellant Engine (LPE)’s feed system. To this end, the new hybrid meta-model methodology by utilizing the Design of Experiment (DOE) method and the Response Surface Method (RSM) were developed and implemented as two effective means of designing, analyzing and optimizing. The input design variables, constraints, objective function, and their surfaces were identified. Then, the design and development strategy was clarified by utilizing the combination of RSM, DOE and regression analysis. Hence, ۶۴ different experiments were carried out on the RD-۲۵۳ propulsion system. The response surface curves were drawn and the related objective function equation was obtained. The Analysis of Variance (ANOVA) results indicate that, the developed hybrid model is capable to predict the responses adequately within the limits of input parameters. In addition, the precision of the model was assessed by comparing with the existing samples and the output was interpreted and analyzed that shown highly accuracy. Therefore, desirability function analysis has been applied to LPE’s feed system for achieving to maximize the power and minimize the weight, simultaneously. Finally, confirmatory tests have been conducted with the optimum parametric conditions to validate the optimization techniques. In conclusion, the methodology capability is to optimize the LPE system, an ۱۱% increase in the power to feed system weight ratio and a ۲% increase the thrust to engine weight ratio. These values are considerably large for LPE design.

کلیدواژه ها:

New Hybrid Meta-Model ، Liquid Propellant Engine (LPE) ، Feed System ، Design of Experiment (DoE) ، Response Surface Method (RSM)