基于MABC-BPNN的内蒙古西部超低能耗草原民居多目标优化预测模型  

MABC-BPNN based multi-objective optimization prediction model for ultra-low energy consumption residential houses in western Inner Mongolia grassland

作　　者：金国辉[1,2] 马骥 杨鹏 陈伟 JIN Guohui;MA Ji;YANG Peng;CHEN Wei(School of Civil Engineering,Inner Mongolia University of Science and Technology,Baotou 014010,China;State Key Laboratory of Green Building in Western China,Xi’an University of Architecture and Technology,Xi’an 710043,China)

机构地区：[1]内蒙古科技大学土木工程学院,包头014010 [2]西安建筑科技大学西部绿色建筑国家重点实验室,西安710043

出　　处：《干旱区资源与环境》2021年第6期73-79,共7页Journal of Arid Land Resources and Environment

基　　金：国家自然科学基金(51768053);内蒙古自然科学基金(2019MS05007);西部绿色建筑国家重点实验室培育基地开放研究基金(LSKF201803)资助。

摘　　要：为解决内蒙古西部草原民居建筑存在的无外围护结构保温设施、采暖方式落后导致的冬季采暖所需能耗量大、热舒适度较差的问题,文中旨在解决内蒙古西部草原民居的痛点问题,利用BP神经网络构建内蒙古西部草原民居建筑超低能耗预测模型,选取多个影响能耗的影响因素,使用Design-Builder软件进行能耗模拟,利用Design-Builder软件所得参数作为基本样本特征,构建基于MABC-BPNN的民居建筑超低能耗预测模型。文中通过利用改进人工蜂群算法,使用BPNN的预测模型作为目标函数,建立内蒙古西部草原民居多目标优化模型,从510组方案中进行优化方案的选取,最终经过迭代计算,选出最优方案,结果表明:通过对内蒙古西部超低能耗草原民居进行多目标优化后,民居的冬季采暖能耗相比降低了67.8%,房屋的外窗以及保温层造价减少了5.4%,通过两者之间的对比发现室内热舒适也提高了,PMV值表现为提升了0.8%,从而达到了优化目的。In order to solve the problems of more energy consumption and poor thermal comfort caused by the lack of thermal insulation facilities and backward heating methods in the grassland residential buildings in western Inner Mongolia,Arming to solve the pain points of the residential buildings in the western grasslands of Inner Mongolia,BP neural network was used to construct an ultra-low energy consumption prediction model for residential buildings in the grasslands of western Inner Mongolia.Multiple factors that affect energy consumption were selected to simulate energy consumption by using Design-Builder software,and the parameters obtained from Design-Builder software as basic sample characteristics were applied to construct a prediction model of ultra-low energy consumption for residential buildings based on MABC-BPNN.By using the improved artificial bee colony algorithm and the BPNN prediction model as the objective function,we established the multi-objective optimization model of grassland dwellings in the west of Inner Mongolia,selected the optimization scheme from 510 groups of schemes,and finally the optimal scheme through iterative calculation was selected.The results show that:after the multi-objective optimization of the ultra-low energy consumption grassland dwellings in the west of Inner Mongolia,the performance of the dwellings is improved.The energy consumption of seasonal heating is reduced by 67.8%,and the cost of external windows and insulation layer is reduced by 5.4%.Through the comparison between the two,it is found that the indoor thermal comfort is also improved,and the PMV value is increased by 0.8%,so as to achieve the optimization purpose.

关 键 词：内蒙古西部 草原民居 超低能耗 改进人工蜂群算法 BP神经网路 

分 类 号：TU253[建筑科学—建筑设计及理论] TU111.4