Sha Song
Low energy office, NSGA2, GANN, multi objective optimisation, architectural design
With the continuous growth of global energy demand and the increasing severity of climate change, seeking innovative solutions to reduce building energy consumption (ECO) has become crucial. This study introduces an integrated algorithm of artificial neural networks (ANN) and limit gradient enhancement, and applies it to the training of building ECO prediction models. Meanwhile, it uses the non dominated sorting genetic algorithm 2 (NSGA2) algorithm for multi-objective optimisation of building design to find the optimal solutions in energy density. The results show that the target parameters of the modified model obtained using the proposed algorithm are superior to the equivalent model. Its annual ECO density is only 25.364 kWh/m2. Meanwhile, its thermal comfort time rate is as high as 0.781, and the incremental cost of the enclosure structure is only 172.08. This indicates that this method can effectively optimise the design of low energy office (LEO) buildings, thereby improving the performance and efficiency of office buildings. This can provide reliable technical support for modern sustainable building design.
Important Links:
Go Back
