基于BIM和SVM的办公建筑冬季采暖能耗预测模型  

Winter Heating Energy Consumption Prediction Model of Office Building Based on BIM and SVM

作　　者：高新琦 GAO Xin-qi(School of Architecture and Environmental Arts,Shanghai Urban Construction College,Shanghai 201415,China)

机构地区：[1]上海城建职业学院建筑与环境艺术学院,上海201415

出　　处：《辽东学院学报（自然科学版）》2023年第1期51-56,共6页Journal of Eastern Liaoning University:Natural Science Edition

基　　金：上海市教委高职院校虚拟仿真基地项目(2021-28)。

摘　　要：为避免办公建筑冬季采暖能耗受墙体保温性能等多因素影响,设计一种基于BIM和SVM的办公建筑冬季采暖能耗预测模型。使用Revit Architecture软件构建办公建筑参数化BIM模型。在建模前设定相关参数,对影响因素实施简化处理。利用Revit构建包括柱、梁、墙等办公建筑的主体参数化模型;利用Revit MEP构建供暖管道系统的参数化模型。利用遗传算法改进SVM算法并构建办公建筑冬季采暖能耗预测模型,实现冬季采暖能耗预测。测试结果表明,设计模型的逐时预测准确性与逐日预测准确性都较高,建模时间在5000 s左右,预测时间较短。In order to avoid the influences such as wall insulation performance on office building heating energy consumption in winter,a prediction model of office building heating energy consumption in winter based on BIM and SVM was designed.The parametric BIM model of office building was constructed by using Revit software.Relevant parameters were set before modeling to simplify the influencing factors.Revit Architecture software was used to build a parametric model of office building main body including columns,beams and walls;and Revit MEP was used to constructed the parametric model of heating pipeline system.Combined with SVM algorithm,the winter heating energy consumption prediction model of office buildings was constructed.The test results show that the hourly prediction accuracy and daily prediction accuracy of the designed model are high,the modeling time is about 5000 s,and the prediction time is short.

关 键 词：BIM SVM 办公建筑 遗传算法 冬季采暖能耗预测模型 

分 类 号：TU111.19[建筑科学—建筑理论]