A dataset of EV batery charging from Shenzhen Auto Electric Power Plant Co., Ltd (Autosun) and Hong Kong Polytechnic University.

A dataset of EV batery in Shenzehn, China.

This dataset covering the period between October 2020 and October 2023 was collected from 30 real public EV charging stations which are operated by Shenzhen Auto Electric Power Plant Co., Ltd (Autosun), a leading company in EV Mega-Watt charging technology solution with a 250 MW public EV charging network in China.

We at first do the preprocessing here including: (1) down-sampling of normal operating data; (2) dividing charging stations into 8 data owners; (3) dropping incomplete or abnormal data. After this, we provide 2 files. The first one (processed_data.xlsx) is the pre-processed data. The second (processed_data_longer_than_30.xlsx) is the data which only includes charging sequences longer than 30 minutes.

The second dataset, which will be used in our article, contains 21175 pieces of EV charging data with 1547432 sampling points, covering 10154 EVs that comprise various types of batteries. Each piece of charging data includes charging current, voltage and power at a 1 minute time interval. The battery fault data in this dataset can be categorized into two distinct types. The first type refers to faults that have been reported by engineering personnel, such as thermal runaway, unbalanced cell voltages, short-circuit and so on. The second type is characterized by excessively high temperatures that incur disconnection by the stations themselves.

The meaning of all columns is shown here:

id: unique id for every data point
transaction_id: unique id for every charging sequence
begin_time: begin time of a charging sequence
end_time: current sampling time
total_charging_kwh: charging energy for now
total_charging_min: charging time for now
current_soc: SOC for now
current_energy_meter_value: the energy of this battery at this time
chargingv: current charging voltage
charginga: current charging current
out_power: current charging power
charging_gun_temperature1: temperature of the charging gun
charging_gun_temperature2: temperature of the charging gun
types: Battery types, 03: LFP, 06: NMC, 04: LMO, 05: LCO, 07: LP
class_judge: data owner index
label: denote if it is a fault battery data, 1 means faulty, 0 means normal

中国深圳的电动汽车（Electric Vehicle, EV）电池数据集。

本数据集的采集时段为2020年10月至2023年10月，数据来源于30座真实公共电动汽车充电站，这些充电站由深圳汽车电力厂有限公司（Autosun）运营——该公司是中国兆瓦级电动汽车充电技术解决方案领域的领军企业，拥有规模达250兆瓦的公共电动汽车充电网络。

我们首先对数据进行了预处理，具体包括：(1) 对正常运行数据进行下采样；(2) 将充电站划分为8个数据持有方；(3) 剔除不完整或异常数据。预处理完成后，我们提供两份数据文件：第一份为`processed_data.xlsx`，即预处理后的数据集；第二份为`processed_data_longer_than_30.xlsx`，仅包含时长超过30分钟的充电序列数据。

本研究将使用的第二份数据集包含21175条电动汽车充电数据，共计1547432个采样点，覆盖10154台搭载不同类型电池的电动汽车。每条充电数据以1分钟为时间间隔，记录了充电电流、电压与功率信息。

本数据集内的电池故障数据可分为两类截然不同的类型：第一类为工程人员上报的故障，例如热失控（thermal runaway）、单体电压不均衡、短路等；第二类则为充电站因温度过高自行触发的断充故障。

各字段含义如下：
`id`：每个数据点的唯一标识符
`transaction_id`：每条充电序列的唯一标识符
`begin_time`：充电序列的开始时间
`end_time`：当前采样时间
`total_charging_kwh`：当前累计充电电量（单位：千瓦时）
`total_charging_min`：当前累计充电时长（单位：分钟）
`current_soc`：当前荷电状态（State of Charge, SOC）
`current_energy_meter_value`：当前时刻电池的累计储能值
`chargingv`：当前充电电压
`charginga`：当前充电电流
`out_power`：当前充电功率
`charging_gun_temperature1`：充电枪1的温度
`charging_gun_temperature2`：充电枪2的温度
`types`：电池类型，编码规则为：03=磷酸铁锂电池（Lithium Iron Phosphate, LFP）、06=镍锰钴三元电池（Nickel Manganese Cobalt, NMC）、04=锰酸锂电池（Lithium Manganese Oxide, LMO）、05=钴酸锂电池（Lithium Cobalt Oxide, LCO）、07=LP型电池
`class_judge`：数据持有方索引
`label`：电池数据的故障标签，1代表故障电池数据，0代表正常电池数据