点击次数：

论文类型：期刊论文

发表时间：2018-12-01

发表刊物：INDOOR AND BUILT ENVIRONMENT

收录刊物：SCIE、Scopus

卷号：27

期号：10

页面范围：1379-1391

ISSN号：1420-326X

关键字：Inverse design; Enclosed environment; Proper orthogonal decomposition; Data interpolation; Aircraft cabin; Computational fluid dynamics

摘要：To design a comfortable aircraft cabin environment, designers conventionally follow an iterative guess-and-correction procedure to determine the air-supply parameters. The conventional method has an extremely low efficiency but does not guarantee an optimal design. This investigation proposed an inverse design method based on a proper orthogonal decomposition of the thermo-flow data provided by full computational fluid dynamics simulations. The orthogonal spatial modes of the thermo-flow fields and corresponding coefficients were firstly extracted. Then, a thermo-flow field was expressed into a linear combination of the spatial modes with their coefficients. The coefficients for each spatial mode are functions of air-supply parameters, which can be interpolated. With a quick map of the cause-effect relationship between the air-supply parameters and the exhibited thermo-flow fields, the optimal air-supply parameters were determined from specific design targets. By setting the percentage of dissatisfied and the predicted mean vote as design targets, the proposed method was implemented for inverse determination of air-supply parameters in two aircraft cabins. The results show that the inverse design using computational fluid dynamics-based proper orthogonal decomposition method is viable. Most of computing time lies in the construction of data samples of thermo-flow fields, while the proper orthogonal decomposition analysis and data interpolation is efficient.