Multi-objective optimization of an endoreversible closed Atkinson cycle
Zheng Gong, Yanlin Ge, Lingen Chen, Huijun Feng
Abstract
Abstract Based on finite-time-thermodynamic theory and the model established in previous literature, the multi-objective optimization analysis for an endoreversible closed Atkinson cycle is conducted through using the NSGA-II algorithm. With the final state point temperature ( T 2 ) of cycle compression process as the optimization variable and the thermal efficiency ( η ), the dimensionless efficient power ( <m:math xmlns:m="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <m:msub> <m:mrow> <m:mover accent="true"> <m:mrow> <m:mi>E</m:mi> </m:mrow> <m:mo>̄</m:mo> </m:mover> </m:mrow> <m:mrow> <m:mi>P</m:mi> </m:mrow> </m:msub> </m:math> ${\bar{E}}_{P}$ ), the dimensionless ecological function ( <m:math xmlns:m="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <m:mrow> <m:mover accent="true"> <m:mrow> <m:mi>E</m:mi> </m:mrow> <m:mo>̄</m:mo> </m:mover> </m:mrow> </m:math> $\bar{E}$ ) and the dimensionless power ( <m:math xmlns:m="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <m:mrow> <m:mover accent="true"> <m:mrow> <m:mi>P</m:mi> </m:mrow> <m:mo>̄</m:mo> </m:mover> </m:mrow> </m:math> $\bar{P}$ ) as the optimization objectives, the influences of T 2 on the four optimization objectives are analyzed, multi-objective optimization analyses of single-, two-, three- and four-objective are conducted, and the optimal cycle optimization objective combination is chosen by using three decision-making methods which include LINMAP, TOPSIS, and Shannon Entropy. The result shows that when four-objective optimization is conducted, with the ascent of T 2 , <m:math xmlns:m="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <m:mrow> <m:mover accent="true"> <m:mrow> <m:mi>P</m:mi> </m:mrow> <m:mo>̄</m:mo> </m:mover> </m:mrow> </m:math> $\bar{P}$ descends, η ascends, both <m:math xmlns:m="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <m:mrow> <m:mover accent="true"> <m:mrow> <m:mi>E</m:mi> </m:mrow> <m:mo>̄</m:mo> </m:mover> </m:mrow> </m:math> $\bar{E}$ and <m:math xmlns:m="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <m:msub> <m:mrow> <m:mover accent="true"> <m:mrow> <m:mi>E</m:mi> </m:mrow> <m:mo>̄</m:mo> </m:mover> </m:mrow> <m:mrow> <m:mi>P</m:mi> </m:mrow> </m:msub> </m:math> ${\bar{E}}_{P}$ firstly ascend and then descend. In this situation, the deviation index is the smallest and equals to 0.2657 under the decision-making method of Shannon Entropy, so its optimization result is the optimal. The multi-objective optimization results are able to provide certain guidelines for the design of practical closed Atkinson cycle heat engine.