ABSTRACT Residential energy models are a common tool for determining the overall energy consumption attributed to the housing sector of a country as well as for projecting the future energy demand in relation to energy conservation policies. However, current residential energy models often require large amounts of input data, have a limited transferability to different countries and often fail to correctly depict trends in energy consumption. Furthermore, no current model gives an indication of the underlying uncertainties in its results. This paper presents a transferable residential energy model that combines statistical and building physics approaches and, with input data typically available in most countries, is able to model the annual and monthly residential energy demand of a given nation or region. In addition to providing trends in the residential sector’s final energy demand according to area of use, the model also estimates uncertainties in the results in three probability bands at 30%, 60% and 90% confidence intervals. Results for a ‘forecast period’ 2001–2010 in three case study countries: Germany, Chile and the UK show a high overall agreement of the new modelling approach with the statistical data and the residential sector’s energy consumption trends of these three countries.

中文翻译：

---

用于确定一个国家住宅部门未来能源需求及其不确定性的可转移能源模型

摘要 住宅能源模型是确定一个国家住房部门的总体能源消耗以及预测与节能政策相关的未来能源需求的常用工具。然而，当前的住宅能源模型通常需要大量输入数据，对不同国家的可移植性有限，并且往往无法正确描绘能源消耗趋势。此外，目前没有任何模型表明其结果中​​存在潜在的不确定性。本文提出了一种可转移的住宅能源模型，该模型结合了统计和建筑物理学方法，并且利用大多数国家/地区通常可获得的输入数据，能够对给定国家或地区的年度和月度住宅能源需求进行建模。除了根据使用区域提供住宅部门最终能源需求的趋势外，该模型还在 30%、60% 和 90% 置信区间的三个概率带中估计结果的不确定性。三个案例研究国家 2001-2010 年“预测期”的结果：德国、智利和英国显示了新建模方法与这三个国家的统计数据和住宅部门能源消耗趋势的高度总体一致。