UNIBO Powertools Dataset

The UNIBO Powertools Dataset has been collected in a laboratory test by an Italian Equipment producer. The cycling experiments are designed to analyze different cells intended for use in various cleaning equipment such as vacuum and automated floor cleaners. The vast dataset is composed of 27 batteries. The main features of the dataset are: (1) the use of batteries from different manufacturers, (2) cells with several nominal capacities, (3) cycling is performed until the cell's end-of-life and thus data regarding the cell at different life stages are produced. Three types of tests have been conducted. (I) The standard test, where the battery was discharged at 5A current in main cycles. (II), the high current test, where the battery was discharged at 8A current in main cycles. (III), the preconditioned test, where the battery cells are stored at 45°C environments for 90 days before conducting the test. During discharge, the sampling period is 10 seconds. The experiments were conducted using the following procedure: (1) Charge cycle: Constant Current-Constant Voltage (CC-CV) at 1.8A and 4.2V (100mA cut-off ) (2) Discharge cycle: Constant Current until cut-off voltage (2.5V) (3) Repeat steps 1 and 2 (main cycle) 100 times (4) Capacity measurement: charge CC-CV 1A 4.2V (100mA cut-off ) and discharge CC 0.1A 2.5V (5) Repeat the above steps until the end of life of the battery cell Data are described in data-description.md test_result.csv contains all the records but the last one of each charge/discharge run. test_result.csv contains the last record of each run. A python file for loading data is available at https://github.com/KeiLongW/battery-state-estimation If you use this dataset, please cite our paper: Kei Long Wong, Michael Bosello, Rita Tse, Carlo Falcomer, Claudio Rossi, Giovanni Pau. 2021. Li-Ion Batteries State-of-Charge Estimation Using Deep LSTM at Various Battery Specifications and Discharge Cycles. In Conference on Information Technology for Social Good (GoodIT ’21), September 9–11, 2021, Roma, Italy. ACM, New York, NY, USA, 7 pages. https://doi.org/10.1145/3462203.3475878

UNIBO Powertools 数据集（UNIBO Powertools Dataset）由意大利一家设备制造商在实验室测试中采集。本循环实验旨在分析适用于各类清洁设备（如真空吸尘器及自动地面清洁机）的多款电池电芯。本数据集共包含27组电池，其核心特征如下：(1) 采用不同厂商生产的电池电芯；(2) 涵盖多种标称容量的电芯；(3) 循环测试持续至电芯寿命终止，因此可获取电芯在不同寿命阶段的相关数据。本次实验共开展三类测试：(I) 标准测试：主循环阶段以5A电流对电池进行放电；(II) 大电流测试：主循环阶段以8A电流对电池进行放电；(III) 预处理测试：在开展正式测试前，将电池电芯置于45℃环境中储存90天。放电过程的采样周期为10秒。实验流程如下：(1) 充电阶段：采用恒流-恒压（Constant Current-Constant Voltage, CC-CV）模式，充电电流为1.8A，充电电压设为4.2V（截止电流100mA）；(2) 放电阶段：采用恒流放电模式，直至电压降至截止值2.5V；(3) 重复步骤1与步骤2（即主循环）共计100次；(4) 容量测量：以恒流-恒压模式以1A电流、4.2V充电（截止电流100mA），并以0.1A恒流放电至2.5V；(5) 重复上述所有步骤，直至电池电芯达到寿命终止状态。相关数据说明详见data-description.md文件。test_result.csv包含每次充放电运行除最后一条记录外的全部数据，test_result.csv包含每次运行的最后一条记录。配套的数据加载Python脚本可通过以下链接获取：https://github.com/KeiLongW/battery-state-estimation。若使用本数据集，请引用以下论文：Kei Long Wong, Michael Bosello, Rita Tse, Carlo Falcomer, Claudio Rossi, Giovanni Pau. 2021. 《基于深度长短期记忆网络的不同规格与放电循环锂离子电池荷电状态估算》. 见：2021年社会公益信息技术会议（GoodIT ’21），2021年9月9日至11日，意大利罗马。美国计算机协会（ACM），纽约州纽约市，共7页。DOI: 10.1145/3462203.3475878