Development of polynomial regression models for composite dynamic envelopes’ thermal performance forecasting

A review on hybrid optimization algorithms to coalesce computational morphogenesis with interactive energy consumption forecasting

“…These simplified models have been used as objective functions, which are the critical required information for a robust GA optimization process, reducing at the same time the complexity of the problem. The results of many researches showed that the coupling of Genetic Algorithms with such simplified established models could rapidly predict and optimize the variation of the heat gain and loss characteristics that was caused by the modification of the overall shape [32,65,67,[75][76][77]. 2.…”

“…Then geometric information can be extracted from each finite element and is used to estimate the heat gain and loss characteristics of the whole architectural shape as well as thermal storage and load shifting qualities of the envelope. Past research achieved to couple the aforementioned computational morphogenesis process with genetic algorithm (GA) free-form building shape optimization for a variety of climate zones [32,[65][66][67]77]. The optimization process is based on a variety of established and sufficiently validated-evaluated reduced models that returned the thermal characteristics of free shapes in an interactive way.…”

“…The polynomial models incorporated in the optimization software resulted from simple factorial and fractional factorial simulation plans, and are conceived in order to forecast a composite dynamic envelope's thermal performance in terms of thermal resistance and daylight availability through the concept of daylight factor. At a first stage, the polynomial regression models link the thermal performance of a composite wall to the inclination angle, the thermal conductivity and the thickness of each envelope's layer [32]. In addition, the authors aimed to enhance software's applicability including energy reduction due to lighting issues via visual comfort optimization.…”

“…Finally, Marsault [68], Marin et al [69] and Mavromatidis et al [32,67], developed EcCoGen, which is a kind of software that belongs to the family of tools based on interactive generative genetic algorithm optimization. The software generates solutions evaluated according to certain criteria, including mainly energy performance and daylight issues.…”

“…Even if the building research community has been always using building performance simulation tools [22][23][24][25][26][27][28][29], this kind of research is not coupled with the architectural design process at early design stages. As it is well-argued elsewhere, there are many reasons for this delay, starting from the difficulty of using complex sophisticated simulation tools to describe complex physical phenomena [26,28,[30][31][32], the augmented calculation time needed for such operations, the uncertainty in the results and the general impression that the designer is restricted by the limitations of the tools [33,34]. Architectural discourse has always included a reflection on providing a high quality and comfortable interior environment via an equilibrated architectural project, albeit the use of computer-aided building energy simulation software in the decision-making process for building form optimization at early conceptual stages is limited.…”