KRESNIK anonymized data and regression models

This repository contains an anonymized version of the KRESNIK database used in the research listed below. KRESNIK contains data from 384 facade tests. Most of these are commercial tests provided by ITB and Arup, with some publicly available tests and experiments included as well. The database includes tests using the PN-B-02867, NFPA 285, and BS 8414 facade test standards. As the PN-B-02867 tests are required to be repeated before issuing a pass or fail, the results of these repeat tests have been averaged in this anonymized version, meaning the total number of facades in the dataset is 171. The repository also includes the regression models used in the paper: "Simple and complex regression models to predict facade fire behaviour from existing test data: interpolating and extrapolating". Different iterations of the KRESNIK database have been used in the following research: Bonner M, Wegrzynski W, Papis BK, Rein G (2020) KRESNIK: A top-down, statistical approach to understand the fire performance of building facades using standard test data. Building and Environment 169:106540. https://doi.org/10.1016/j.buildenv.2019.106540 Bonner M (2020) A top-down approach to understand the fire performance of building facades using standard test data. Imperial College London. https://doi.org/10.25560/91502本仓库收录了下述研究中使用的KRESNIK数据库的匿名化版本。KRESNIK数据库包含384次外墙测试的数据，其中绝大多数为由ITB与奥雅纳（Arup）提供的商业测试数据，同时也纳入了部分公开可获取的测试与实验数据。该数据库涵盖遵循PN-B-02867、NFPA 286及BS 8414外墙测试标准的测试项目。由于PN-B-02867标准要求在给出合格或不合格结论前需重复开展测试，本匿名化版本已将这些重复测试的结果取平均值，因此本数据集中实际的外墙样本总数为171。 本仓库同时收录了发表于论文《基于既有测试数据预测外墙防火性能的简单与复杂回归模型：内插与外推》（"Simple and complex regression models to predict facade fire behaviour from existing test data: interpolating and extrapolating"）中所使用的回归模型。 不同迭代版本的KRESNIK数据库已被应用于下述研究： Bonner M、Wegrzynski W、Papis BK、Rein G（2020）《KRESNIK：基于标准测试数据解析建筑外墙防火性能的自上而下统计方法》，《建筑与环境（Building and Environment）》，第169卷：106540。https://doi.org/10.1016/j.buildenv.2019.106540 Bonner M（2020）《基于标准测试数据解析建筑外墙防火性能的自上而下方法》，帝国理工学院（Imperial College London）。https://doi.org/10.25560/91502