Model for: A high-temporal resolution residential building occupancy model to generate high-temporal resolution heating load profiles of occupancy-integrated archetypes

The model is capable of creating stochastic multi-day occupancy profiles for building stock of different sizes and characterised by different shares of households belonging to the different occupancy categories identified in the UK. The model uses the Monte Carlo Markov Chain technique. The occupancy categories are developed by the application of a data-mining clustering technique on data available from the UK Time Use Survey 2015. These categories are characterised by the following occupancy profiles: 1. Daily absence: unoccupied period from 09.00 to 04:00, 2. Working hours absence: unoccupied period from 08:20 to 18:10, 3. Lunchtime absence: unoccupied period from 11:10 to 16:10, 4. Constant presence 1, 5. Constant presence 2. These occupancy categories are described in details in the associated paper and in a previous publication (https://doi.org/10.1016/j.enbuild.2019.05.056.). In the associated paper the stochastic occupancy profiles are used as inputs in energy models of residential buildings, but the source code may be readily adapted for specific applications, with due acknowledgement to the authors.本模型可针对不同规模的建筑存量生成随机多日居住占用剖面，而此类建筑存量的特征为：家庭按英国已识别的各类居住类别划分的占比各有不同。该模型采用蒙特卡洛马尔可夫链（Monte Carlo Markov Chain）技术。 上述居住类别基于2015年英国时间使用调查（UK Time Use Survey 2015）的数据，通过数据挖掘聚类技术构建得到。各类居住类别对应如下居住占用剖面特征： 1. 全日空置：每日09:00至次日04:00处于无人居住状态； 2. 工作时段空置：每日08:20至18:10处于无人居住状态； 3. 午餐时段空置：每日11:10至16:10处于无人居住状态； 4. 持续居住状态1； 5. 持续居住状态2。 上述居住类别的详细说明可参阅相关研究论文及此前发表的文献（https://doi.org/10.1016/j.enbuild.2019.05.056.）。 在相关研究中，该随机居住占用剖面被用作住宅建筑能源模型的输入参数，但本源代码可在恰当引用作者的前提下，灵活适配各类特定应用场景。