| In recent years,with the continuous improvement of the living standards of grassland herdsmen,people’s requirements for indoor thermal environment in winter have gradually increased.Nowadays,there are many problems in grassland residential buildings,such as rough building structure,poor thermal insulation performance and weak awareness of energy saving of herdsmen,which leads to high heating energy consumption and poor indoor thermal comfort of grassland residential buildings in winter.Therefore,this paper takes the energy consumption of building heating,indoor thermal comfort,the engineering cost as the optimization objectives,and establishes a multi-objective optimization model of building energy saving,which provides a new idea for the research of grassland residential buildings in western Inner Mongolia.The optimization model has guiding significance for the optimal design and construction of grassland residential buildings in western Inner Mongolia.This paper proposes a multi-objective optimization model for building energy conservation by means of literature analysis,data analysis,quantitative and qualitative analysis,and interdisciplinary research.By analyzing the correlation between the influencing factors and the optimization goal of building energy saving,the selected eastfacing window-to-wall ratio,the west-facing window-to-wall ratio,the south-facing window-to-wall ratio,the north-facing window-to-wall ratio,the east outer window glass heat transfer coefficient,and the west the heat transfer coefficient of the window glass,the heat transfer coefficient of the south outer window glass,the heat transfer coefficient of the north outer window glass,the heat transfer coefficient of the outer wall and the heat transfer coefficient of the roof are 10 influencing factors,and these factors are used as the optimization variables.The building heating energy consumption,indoor thermal comfort PMV,and engineering cost were selected as three optimization targets.Firstly,through the different values of 10 optimization variables,the training sample design is carried out,and the results of the simulation calculation are used as the training samples and test samples of the BPNN prediction model.The average error value of the test sample of BPNN prediction model is calculated to be 0.018 6,which verifies the feasibility of the prediction model.Secondly,The trained BPNN prediction model is used as the fitness function of multiobjective optimization.The ABC algorithm is used for multi-objective optimization,and the ABC-BPNN multi-objective optimization model for energy saving in the grassland residential buildings in Inner Mongolia is obtained.In the scope of building energy-saving design standards,the Pareto optimal solution set of global search optimization variables,the building heating energy consumption decreased by 15.48%,the indoor thermal comfort PMV increased by 1.55%,and the cost of external windows and insulation engineering decreased by 1.52%.Achieve the purpose of optimization.Finally,the effectiveness and practicability of the multi-objective optimization model are verified by empirical analysis.The existing grassland residential buildings in Sumutu Village,Bayanhaote Town,Alashan Zuoqi,western Inner Mongolia were selected for empirical analysis.The ABC-BPNN algorithm and GA-BPNN algorithm were used for multi-objective optimization.The results showed that ABC-BPNN convergence well,the quality of the Pareto optimal solution set is high. |
