Future Weather Files (under Climate Change Scenarios) to Support Building Energy Simulation

The future weather files are synthesized based on the methodology presented in the research project on the impact of climate change on building energy performance. These weather files (in EPW format) are intended to be used for building energy simulation and performance evaluation. The team provides here future weather files for 30 years (from 2020 to 2049) for four Representative Concentration Pathways (RCP2.6, 4.5, 6.0, and 8.5) used in the Fifth IPCC (Intergovernmental Panel on Climate Change) Assessment. Due to limited resources, Canada - Montreal (WMO 716270) is the first set of data available. Weather files for other locations and scenarios will gradually be added. If you have any pressing needs, you are welcome to contact the team to explore the possibility. The future weather files are synthesized based on a workflow developed by the team. A full explanation is presented in our article. The following is a brief description of the workflow: The effect of climate change is based on the General Circulation Models (GCM). First, a bias-correction technique, known as the quantile-quantile method, is applied to remove the bias in the data in order to adopt GCMs to a specific location. A hybrid classification-regression model is used to downscale the bias-corrected GCM data to synthesize future weather data at an hourly resolution for building energy simulation. The proposed workflow enables users to use a set of observed weather data by finding similar patterns rather than artificially generating data. However, in cases where the future GCM data show temperatures outside the range of the observed data, a trained regression model is applied to synthesize hourly weather data. If readers are interested in the topic, please browse through related articles of the team: Generating future weather files under climate change scenarios to support building energy simulation — a machine learning approach https://doi.org/10.1016/j.enbuild.2020.110543 A systematic approach in constructing typical meteorological year weather files using machine learning https://doi.org/10.1016/j.enbuild.2020.110375 Cooling and heating energy performance of a building with a variety of roof designs; the effects of future weather data in a cold climate https://doi.org/10.1016/j.jobe.2018.02.001 Energy performance of cool roofs under the impact of actual weather data https://doi.org/10.1016/j.enbuild.2017.04.006 Version History: Version 1: WMO 716270, RCP 4.5 only Version 2: WMO 716270, RCP 2.6, 6.0, 8.5 added Disclaimer: The weather files provided here are synthesized based on the novel workflow developed by the research team with limitations and based on assumptions. A full description of the workflow is presented in our articles. We do not make any warranties about the completeness, reliability, and accuracy of the data. Any use of the data is strictly at your own risk, and we will not be liable for any losses and damages in connection with the use.

本数据集的未来气象文件基于气候变化对建筑能耗性能影响相关研究项目中提出的方法合成得到。此类采用EPW格式（EnergyPlus气象文件）的气象文件，旨在用于建筑能耗模拟与性能评估工作。

本团队本次发布的未来气象数据涵盖2020至2049年共30年的时长，覆盖政府间气候变化专门委员会（Intergovernmental Panel on Climate Change，IPCC）第五次评估报告所采用的四类典型浓度路径（Representative Concentration Pathways，RCP2.6、4.5、6.0及8.5）。受限于资源条件，本次首批公开的数据为加拿大蒙特利尔（WMO 716270）站点的气象文件。后续将逐步新增更多地区与情景下的气象文件，若您有紧急需求，欢迎联系本团队协商探讨合作可能。

本数据集的未来气象文件基于本团队自研的工作流合成，完整的方法说明已发表于团队的学术论文，以下为该工作流的简要介绍：
气候变化影响分析基于大气环流模式（General Circulation Models，GCM）。首先，通过分位数-分位数（quantile-quantile）偏差校正技术对数据进行偏差校正，以使GCM结果适配特定站点的气象特征。随后，采用混合分类回归模型对校正后的GCM数据进行降尺度处理，以合成适用于建筑能耗模拟的逐小时分辨率未来气象数据。该工作流允许用户通过匹配相似模式调用实测气象数据集，而非人工生成气象数据。但若未来GCM数据中的温度超出实测数据的温度区间，则将调用预训练的回归模型合成逐小时气象数据。

若您对该主题感兴趣，可参阅本团队的相关学术论文：
1. 《气候变化情景下支撑建筑能耗模拟的未来气象文件生成：一种机器学习方法》https://doi.org/10.1016/j.enbuild.2020.110543
2. 《基于机器学习构建典型气象年文件的系统性方法》https://doi.org/10.1016/j.enbuild.2020.110375
3. 《不同屋顶设计建筑的冷热能耗性能：寒冷气候下未来气象数据的影响》https://doi.org/10.1016/j.jobe.2018.02.001
4. 《实际气象数据影响下冷屋顶的能耗性能》https://doi.org/10.1016/j.enbuild.2017.04.006

版本历史：
版本1：仅包含WMO 716270站点的RCP 4.5情景数据
版本2：新增WMO 716270站点的RCP 2.6、6.0及8.5情景数据

免责声明：
本数据集提供的气象文件基于本团队研发的创新性工作流合成，存在一定局限性与假设前提，完整的工作流说明已发表于团队相关论文。本团队不对数据的完整性、可靠性与准确性作出任何明示或默示的保证。任何数据使用行为均由使用者自行承担风险，本团队不对因使用该数据导致的任何损失或损害承担责任。