Short-term Load Forecasting Model with Predicted Weather Data

摘要：

　　Introduced in this paper are our recommended weather forecasting model and the associated prediction procedure.These facilitate the formulation of a convenient and accurate short term energy consumption forecasting method.We started with an overview of the building related energy forecasting.A simplified weather model was then proposed for the output of key weather data,including dry-bulb temperature,relative humidity and global solar radiation.On top of this,a framework of energy-use forecasting technique is introduced,which brings in the data filtering and raw data set regrouping methods.Our result comparisons show that the predictions of the Mean Absolute Percentage Error(MAPE)from the weather data on record,or from the model-generated data set,differ very little.Illustrating examples via a university building case are discussed.The above indicate that our proposed weather prediction model is convenient and suitable for the mentioned purposes.Comparisons among the different input data sets also demonstrate that the forecasting accuracy has been enhanced through input data filtering and regrouping.

关键词： weather prediction energy consumption forecasting energy saving

作者: ZHU Guangya CHOW Tin-tai Norman TSE

作者单位: City university of Hong Kong,Hong Kong SAR

会议类型: 国际会议

会议名称: 2017世界可持续建筑环境大会(the World Sustainable Built Environment Conference)

会议地点: 香港

会议语种:英文

页码: 1748-1754

在线出版日期: 2017-06-05（万方平台首次上网日期，不代表论文的发表时间）