摘要
针对建筑能耗预测中单一模型精度较低的问题,提出了基于多种神经网络组合的建筑预测模型。首先对影响建筑物能耗的主要天气因素分析,并使用人体舒适度指数优化部分数据。再针对GA-BP神经网络、RBF神经网络、广义回归神经网络,使用等权重法和优势矩阵法两种组合方法,建立组合预测模型。以某图书馆的历史数据作为实际算例,将两种组合模型与四种单一模型的结果进行比较。仿真结果表明,所提方法相比于传统方法有更高的精确度。
According to the problem of low accuracy of a single model in building energy consumption prediction,ensemble prediction models based on three artificial neural networks are proposed.First,the main weather factors that affect building energy consumption were analyzed,and the human body amenity indicator was used to optimize some data.Secondly,for GA-BP neural network,RBF neural network,and generalized regression neural network,two methods of equal weight and odds-matrix were used to establish ensemble prediction models.Taking the historical data of a library as an actual example,the results of the two ensemble methods and other single neural networks are compared.The results show that the methods proposed in the article have higher prediction accuracy than traditional models.
作者
窦嘉铭
马鸿雁
DOU Jia-ming;MA Hong-yan(School of Electrical and Information Engineering,Beijing University of Civil Engineering and Architecture,Beijing 100044,China;National Virtual Simulation Experimental Center for Smart City Education,Beijing 100044,China;Beijing Key Laboratory of Intelligent Processing for Building Big Data,Beijing 100044,China)
出处
《计算机仿真》
北大核心
2022年第5期438-443,共6页
Computer Simulation
基金
北京建筑大学博士基金项目(ZF15054)
北京建筑大学研究生创新项目(PG2020050)。
关键词
建筑能耗
组合模型
神经网络
Building energy consumption
Ensemble models
Neural network
