Single-, two-, three-, and four-objective optimizations for an irreversible vacuum thermionic generator via finite-time thermodynamics, NSGA-II and three decision-making techniques

Abstract Based on an irreversible vacuum thermionic generator model, this paper performs single-, two-, three-, and four-objective optimizations for it by utilizing NSGA-II algorithm, finite-time thermodynamics and three decision-making techniques. Output-power, thermal-efficiency, ecological-function and efficient-power are taken as optimization-objectives, and collector work-function and output-voltage are taken as optimization-variables. Total number of optimizations completed consists of one four-objective optimization, four three-objective optimizations, six two-objective optimizations, and four single-objective optimizations. Deviation indexes under various combinations of optimization objectives are compared by using three decision-making techniques: Shannon Entropy, LINMAP, and TOPSIS. Based on these comparisons, optimal design scheme is determined. Findings indicate that multi-objective optimization can take into account multiple performance indicators at the same time and have a better coordination ability. The deviation index is 0.1761 when selecting LINMAP decision mode to optimize the combination of output-power, efficient-power, thermal-efficiency and ecological-function which is the best scheme for four-objective optimization, and optimal ranges of collector work function and output voltage are 1.295–1.355 and 0.22–0.44, respectively. The important contribution herein is introducing NSGA-II algorithm and three decision-making techniques for performance optimization of vacuum thermionic generator and performing totally 15 optimization tasks for in the first time in open literature.