Buildings, Год журнала: 2024, Номер 14(1), С. 284 - 284

Опубликована: Янв. 20, 2024

The thermal comfort evaluation of the urban environment arouses widespread concern among scholars, and research in this field is mostly based on indexes such as PMV, PET, SET, UTCI, etc. These index models are complex calculation process poor operability, which makes it difficult for people who lack a relevant knowledge background to understand, calculate, apply them. purpose study provide simple, efficient, easy-to-operate outdoor model severe cold areas China using machine learning method. In study, physical parameters obtained by measurement, individual information questionnaire survey. applicability four studied. A total 320 questionnaires collected. results show that correlation coefficients between predicted values voting extreme gradient lifting model, random forest neural network 0.9313, 0.7148, 0.9115, 0.5325, respectively. Further analysis with highest coefficient shows factors (such residence time, distance hometown residence, clothing, age, height, weight) environmental air humidity (RH), wind speed (v), temperature (Ta), black bulb (Tg)) have different influences evaluation. summary, method evaluate simpler, more direct, can make up consideration index. reference value application China.

Язык: Английский