响应面法的温差发电器优化设计  

Optimization Design of Thermoelectric Generator Based on Response Surface Method

作　　者：何嘉华 周宏甫[1] 张征[1] HE Jia-hua;ZHOU Hong-fu;ZHANG Zheng(South China University of Technology,Guangdong Guangzhou 510641,China)

机构地区：[1]华南理工大学,广东广州510641

出　　处：《机械设计与制造》2018年第5期44-47,共4页Machinery Design & Manufacture

基　　金：国家自然科学基金项目(51375168)

摘　　要：为提高温差发电器的发电效率,同时保证整机轻量化,提出了以发电器结构参数为设计变量,以质量、排气背压和发电片两侧温度差为目标函数的整机多目标优化设计方案。首先利用中心复合设计法、有限元法分别进行设计点的选取与计算。其次利用响应面法构建设计变量与质量、温差、排气背压三个响应量的响应面模型,并进行了灵敏度分析。通过多目标遗传优化算法进行基于响应面模型的多目标优化,优化后温差发电器的发电效率提高了7.3%。最后对优化结果进行了可靠性验证。In order to improve power generation efficiency of thermoelectric generator,while ensuring the light weight of it,a multi-objective optimization design scheme was proposed,the structural parameters of the thermoelectric generator as design variables,themass,exhaust back pressure and temperature difference on both sides of thermoelectric module as the objective function of the scheme. Firstly,central composite design and finite element method were used to select and calculate the design points. Secondly,response surface methodology was used to construct the response surface model of design variables tomass, temperature difference, exhaust back pressure. And sensitivity analysis was made. Thirdly, multi-objective optimization based on response surface model was made by using multi-objective genetic algorithm. Power generation efficiency of thermoelectric generator was increased by 7.3% after optimization. Last,the optimization resultswere calculated to verify the reliability.

关 键 词：温差发电器 响应面法 多目标优化 遗传算法 

分 类 号：TH16[机械工程—机械制造及自动化] TK124[动力工程及工程热物理—工程热物理]