基于品牌维度的充电桩充电过程异常原因识别数据

本数据集基于充电桩设备品牌（生产厂家）维度，结合设备近期异常记录，识别充电过程中的典型故障场景与潜在根因。具体应用场景如下：
1.对平台（即申请人）而言：可基于不同厂家设备的故障归因特征，建立品牌级充电过程异常知识库，辅助智能运维模型训练与品牌性能评估，提升平台对接设备商的技术管理能力；
2.对场站商家而言：可借助品牌维度的异常原因归类结果，识别当前场站中高频故障品牌与对应问题模式，指导站点层面的设备替换决策和保修策略，降低维护成本与停机风险；
3.对政府而言：可用于监测市场主流品牌充电设备的运行稳定性，评估不同厂家设备在公共设施中的适配度和可靠性，为政府集中采购目录优化和产品准入制定提供数据依据。1.数据采集：原始数据经授权合法获取，实时采集充电过程异常信息，包括设备生产厂家名称、设备编号、订单编号、异常码、异常时间点、异常码得分、异常码对应故障场景等字段。
2.订单回溯取值：以异常时间点为基准，向前获取该充电设备的最近15笔订单（包含本单），提取每笔订单对应的异常码、该次异常码得分、该次异常码对应故障场景等信息（以[订单编号,异常码,异常码得分,故障场景]格式聚合形成待归类数据集）。
3.异常码归类：依据异常码与故障场景的映射规则，将15笔订单中的异常码分别归入不同的故障场景，并对每个场景下的异常码得分进行累加（以{"场景名称":["订单号_异常码_该次异常码得分",...,"总得分_分值"]}格式输出归类后的待诊断信息）；
4.故障诊断判定：若某一故障场景下累计异常码得分大于等于1，则输出该场景的故障结论。若无场景满足阈值，则不输出结论。

This dataset identifies typical fault scenarios and potential root causes during the charging process based on the brand (manufacturer) dimension of electric vehicle (EV) charging pile equipment, combined with recent abnormal records of the equipment. The specific application scenarios are as follows:
1. For the platform (i.e., the applicant): It can establish a brand-level charging process abnormality knowledge base based on the fault attribution characteristics of equipment from different manufacturers, assist in the training of intelligent operation and maintenance models and brand performance evaluation, and enhance the platform's technical management capability when collaborating with equipment suppliers;
2. For charging station operators: They can use the brand-based abnormal cause classification results to identify high-frequency faulty brands and their corresponding problem patterns at their current stations, guide equipment replacement decisions and warranty strategies at the station level, and reduce maintenance costs and downtime risks;
3. For the government: It can be used to monitor the operational stability of mainstream EV charging equipment brands in the market, evaluate the adaptability and reliability of equipment from different manufacturers in public facilities, and provide data support for the optimization of government centralized procurement catalogs and the formulation of product access standards.

1. Data Collection: The original data is legally obtained with authorization, and real-time collection of abnormal information during the charging process is carried out, including fields such as equipment manufacturer name, equipment serial number, order number, abnormal code, abnormal time point, abnormal code score, and fault scenario corresponding to the abnormal code.
2. Order Backtracking and Value Extraction: Taking the abnormal time point as the benchmark, obtain the latest 15 orders (including the current order) of the charging equipment up to the abnormal time point, extract the abnormal code, the abnormal code score of each order, and the fault scenario corresponding to the abnormal code for each order, and aggregate the information into a dataset to be classified in the format of [order number, abnormal code, abnormal code score, fault scenario].
3. Abnormal Code Classification: According to the mapping rules between abnormal codes and fault scenarios, classify the abnormal codes in the 15 orders into different fault scenarios respectively, and accumulate the abnormal code scores under each scenario. Output the classified diagnostic information in the format of {"scene_name": ["order_id_abnormal_code_this_abnormal_code_score", ..., "total_score_score_value"]}.
4. Fault Diagnosis and Judgment: If the cumulative abnormal code score of a certain fault scenario is greater than or equal to 1, output the fault conclusion of this scenario. If no scenario meets the threshold, no conclusion will be output.

搜集汇总

背景与挑战

背景概述

该数据集由浙江小桔绿色能源科技有限公司登记，包含501条企业数据，以xlsx格式存储充电桩充电过程的异常记录，关键字段包括设备生产厂家名称、异常码和故障场景。其特点在于基于品牌维度识别异常原因，通过算法规则进行故障诊断，适用于平台运维优化、场站设备管理和政府监管评估等多场景应用。