储能电池异常点识别及其精度验证支撑数据集

本数据集由北京理工大学构建，旨在支撑储能电池异常诊断算法的精度验证与性能评估。数据集涵盖五类典型的异常/故障特征数据：1. 析锂异常数据：72块析锂电池与130块正常电池的提取特征（含容量增量曲线、等效电路参数）；2. 内阻异常数据：202块电池的直流内阻（DCR）分布数据，用于DCR一致性与异常诊断；3. 温度异常数据：储能电站环境下64块电池模组的长时间运行温度监测数据；4. 电压异常数据：8块电池在实验室环境下的长时间运行电压波动数据；5. 内短路异常数据：8块电池在内短路发生后的静置电压骤降数据。该数据集融合了实验室测试与电站实测数据，特征明确，适用于开发和验证基于物理模型或数据驱动的电池故障诊断方法。

This dataset was constructed by Beijing Institute of Technology to support the accuracy verification and performance evaluation of anomaly diagnosis algorithms for energy storage batteries. This dataset covers five typical types of anomaly/fault feature data:
1. Lithium plating anomaly data: Extracted features of 72 lithium-plated batteries and 130 normal batteries, including capacity increment curves and equivalent circuit parameters;
2. Internal resistance anomaly data: Distribution data of direct current resistance (DCR) of 202 batteries, used for DCR consistency assessment and anomaly diagnosis;
3. Temperature anomaly data: Long-term operating temperature monitoring data of 64 battery modules in an energy storage power station environment;
4. Voltage anomaly data: Long-term operating voltage fluctuation data of 8 batteries in a laboratory environment;
5. Internal short circuit anomaly data: Sudden voltage drop data of 8 batteries during the resting period after internal short circuit occurs.
This dataset combines laboratory test data and on-site measured data from energy storage power stations, with clear features, and is suitable for developing and validating battery fault diagnosis methods based on physical models or data-driven approaches.

搜集汇总

背景与挑战

背景概述

该数据集由北京理工大学构建，旨在支撑储能电池异常诊断算法的精度验证与性能评估。它涵盖五类典型的异常/故障特征数据，包括析锂、内阻、温度、电压和内短路异常，融合了实验室测试与电站实测数据，特征明确，适用于开发和验证基于物理模型或数据驱动的电池故障诊断方法。