Hourly electricity load profiles of paper producing and food processing industries

The data provided are synthetic hourly electricity load profiles for the paper and food industries for one year. The data have been synthetized from two years of measured data from industries in Chile using a comprehensive clustering analysis. The synthetic data possess the same statistical characteristics as the measured data but are provided normalized to one kW and anonymized in order to be used without confidentiality issues. Three CSV files are provided: food_i.csv, paper_i_small.csv and paper_i_large.csv containing the data of a small food processing industry, a small paper industry, and a medium-large paper industry, respectively. All the three files contain seven columns of data: weekday, month, hour, cluster, min, max, mean. The four first columns index the data in the following way: Month: it includes the range of integer values between 1 and 12 accounting for the consecutive calendar months of a year starting in January (1) and ending in December (12). Weekday: this column has integer values in the range 1 to 7 that are equivalent to the consecutive days of the week starting on Monday (1) and ending on Sunday (7). Hour: it consist of integer values ranging between 1 and 24, which describe the hours of a day. Cluster: The column “cluster” represents the cluster to which this data is associated to. The number of clusters is different for each load profile, as well as the number of days included in each cluster. Since the cluster were calculated for days, a cluster number covers 24 consecutive points of data. The load profile data are provided in the three different columns: min, max and mean: Min: this column provides the min value of the cluster at that time of the day. Therefore, it represents the minimum demand of electricity recorded in all the days belonging to this representative group of data. Max. This column provides the maximum electric load of the cluster at that time of the day. It represents the maximum demand of electricity in all the days belonging to this representative group of data at that hour of the day. Mean: This column provides the average electric load of the cluster at that time of the day. It represents the mean demand for electricity belonging to this representative group of data at that hour of the day. The min, max and mean values are different for each hour of the day. All values are provided in values from 0 to 1 with the unit kW. For details on the clustering procedure or the data itself please refer to the associated paper published in the journal Energy and the one published in Data in Brief journal. The study was supported by the German Federal Ministry of Education and Research - BMBF and the Chilean National Commission for Scientific Research and Technology - CONICYT (grant number BMBF150075) , the Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH through the Energy Program in Chile, and the European Research Council (“reFUEL” ERC-2017-STG 758149).本数据集包含针对智利食品与造纸工业的年度逐小时电力负荷合成曲线。该数据集基于智利工业两年实测电力数据，通过全面的聚类分析（clustering analysis）合成而来。合成数据与实测数据具备一致的统计特征，已归一化至1千瓦（kW）并完成匿名化处理，可无保密顾虑地使用。本次提供三份CSV格式文件：food_i.csv、paper_i_small.csv与paper_i_large.csv，分别对应小型食品加工厂、小型造纸厂以及中大型造纸厂的电力负荷数据。三份文件均包含七列数据：工作日（weekday）、月份（month）、小时（hour）、簇（cluster）、最小值（min）、最大值（max）与平均值（mean）。前四列用于索引数据，具体说明如下：1. 月份（Month）：取值范围为1至12的整数，对应一年中自1月（1）至12月（12）的连续自然月份；2. 工作日（Weekday）：取值范围为1至7的整数，代表一周内自周一（1）至周日（7）的连续日期；3. 小时（Hour）：取值范围为1至24的整数，对应一日内的各个小时；4. 簇（Cluster）：该列表示当前数据所属的簇。不同负荷曲线的簇数量各不相同，每个簇包含的天数也存在差异。由于簇是基于单日数据计算得到，因此一个簇编号对应24个连续的负荷数据点。负荷曲线数据存储于三列中：- 最小值（Min）：该列给出对应时刻该簇的最小电力负荷值，即该代表性数据组所有对应日期中记录到的最低电力需求；- 最大值（Max）：该列给出对应时刻该簇的最大电力负荷值，即该代表性数据组所有对应日期中该小时的最高电力需求；- 平均值（Mean）：该列给出对应时刻该簇的平均电力负荷值，即该代表性数据组所有对应日期中该小时的平均电力需求。各小时的最小值、最大值与平均值均不相同，所有数值的取值范围为0至1，单位为千瓦（kW）。如需了解聚类流程或数据集详情，请参阅发表于《Energy》期刊与《Data in Brief》期刊的相关论文。本研究得到德国联邦教育与研究部（BMBF）及智利国家科学研究与技术委员会（CONICYT，资助编号BMBF150075）、德国国际合作机构（GIZ） GmbH智利能源项目，以及欧洲研究理事会（"reFUEL" ERC-2017-STG 758149）的支持。