A novel data-based control strategy of dynamic thermal comfort for inhabited environment

摘要: Dynamic thermal comfort control can effectively improve the levels of comfortable feeling, energy-saving, and human health. Traditional theories methods based on accurate physical model not solve problems met during analyzing controlling because difficulties in obtaining a environment. This paper presented novel strategy dynamic using online offline dada collected Family members' fuzzy sensations are involved closed loop. A six-input one-output CMAC neural network was used to learn relationship between variables parameters Predicted Mean Vote (PMV). functions as predictive PMV model. Another three-input learned nonlinear temperature decrease humidity deduction summer, when changes value. We also defined concepts personal zones, proposing information fusion method sensation set theory. Procedure computational experiments with these data-based models above designed determine appropriate points order for value periodically be within desired zone or zone. The given help researchers design system detail.