摘要
建立了基于支持向量机(SVM)理论的建筑物空调负荷预测模型。对广州地区某办公楼夏季不同月份的逐时空调负荷,分别用SVM模型和BP神经网络模型进行了训练和预测。仿真结果表明,SVM模型具有更高的预测精度和更好的泛化能力,是建筑物空调负荷预测的一种有效方法。
Based on the theory of support vector machine (SVM), establishes a prediction model for building air conditioning load. An SVM model and back-propagation (BP) neural network model are both used for the hourly air conditioning load prediction of an office building in summer months in Guangzhou area. The simulation results show that the SVM model shows better accuracy and generalization ability, and is effective for building air conditioning load prediction.
出处
《暖通空调》
北大核心
2008年第1期14-18,120,共6页
Heating Ventilating & Air Conditioning
基金
国家自然科学基金资助重点项目(编号:50538040)
国家留学基金(编号:留金出[2006]3037号)
关键词
空调负荷
预测
支持向量机
BP神经网络
air conditioning load, prediction, support vector machine, back-propagation neural network