大赛创新组竞赛赛题发布，金奖10万花落谁家
大赛组委会公布了大赛创新组竞赛赛题，共2道：
1.电动汽车行驶里程预测
2.电动汽车行驶SOC预测创新组赛题基于新能源汽车国家大数据联盟提供的新能源汽车行业车辆运行大数据，围绕新能源汽车电池能量预测、行驶里程统计等方向进行赛题设计，进行数据算法分析计算。赛制分为初赛、复赛和线下决赛，创新组竞赛将在2019年10月启动数据提交与排名通道。
2019年大赛自报名以来就受到广泛关注，为鼓励同学们踊跃参赛，以达到选拔创新人才的宗旨，创新组每道赛题都将产生1名金奖，奖金高达10万元。2019年度赛事大赛组委会提供的现金奖励除创新组奖金和创业组奖金外，还设置了最具商业潜力奖、最佳创意奖、最佳方案奖、最佳人气奖、优秀学校组织奖、最受欢迎企业奖等奖项，总奖金额度不低于60万元。到底谁能够突破重围，摘得桂冠呢？让我们拭目以待。
赛题详情
电动汽车行驶里程预测
1.竞赛题目 Question
电动汽车行驶里程预测
Prediction of electric vehicle driving distance
2.背景介绍 Background
续驶里程作为电动汽车重要的性能参数之一，近年来引起了广泛关注。电动汽车某段时间内的实际行驶里程与动力电池衰退情况、行驶过程的特征以及环境因素相关。准确预测电动汽车某一工况下的行驶里程，可以增强驾驶者对车辆续驶里程的信心，提高能量利用率，在能量管理，出行决策制定等方面具有重要意义。
Drivingrange, a key performance parameter of electric vehicles, has raised greatconcern recent years. The actual driving distance in a certain time period of EVsis related to the degradation of power battery, the characteristics of thedriving process and environmental factors. Accurately predicting the drivingdistance of EVs in a specific condition can increase driver’s confidence indriving range, improve energy utilization efficiency, and is of greatsignificance to energy management, travel decision-making and so on.
3.问题描述 Description of the question
行驶里程受电池能量和行驶能耗等多方面因素影响。参赛者需根据给定数据，综合分析环境状况、能耗、电池衰退情况等因素，合理建立数据模型，预测某一行驶片段的行驶里程。
Drivingdistance is influenced by many factors, such as battery energy and drivingenergy consumption. Contestants are required to analyze environmental conditions,energy consumption, battery degradation and other factors according to thegiven data, and establish an appropriate data model to predict the drivingdistance of a certain driving segment.
4.数据说明 Data description
本赛题数据来源于市面上某款电动汽车的实际运行数据，训练样本和测试样本数据由同一车型的不同车辆产生，所有车辆运行地区均为北京。
训练样本包含5辆电动汽车的实际运行数据（含行驶和充电等状态），数据按时间顺序排列成行，列名见表1。原始数据集中可能存在异常数据，请参赛者自行识别。
注：由于车载BMS和电机控制器采集及传输数据的异步性，可能导致电池组相关数据项和电机相关数据项无法完全匹配。
Thereal-world driving data of electric vehicles are used in this question, and thetraining data and test data are produced by different vehicles with the samemodel, all the vehicles are operated in Beijing.
The training data consist ofdata of 5 vehicles (both driving state and charging state are included), andthe data are arranged in time order in rows. The data items are listed in Table1. Abnormal data may exist in the raw data and need to be identified by thecontestants themselves.
Note: Due to the asynchronybetween BMS and motor controller in collecting and transmitting, the data aboutbattery pack and motor could not be completely matched.
测试样本由15辆车的100个行驶片段构成，0-4号车每辆车10个行驶片段，5-14号车每辆车5个行驶片段，测试样本中的0-4号车与训练样本中的0-4号车一一对应，5-14号车没有对应的训练样本数据。测试样本数据按时间顺序排列成行，列名见表2。数据可能存在异常，需参赛者自行识别。
Test data consist of 100 driving fragments of 15 vehicles, the vehicles No.0-4 provides 10 driving fragments each, while the vehicles No.5-14 provides 5 driving fragments each. The vehicles No.0-4 in test data are corresponding to the vehicles No.0-4 in the training data, and vehicles No.5-14 have no corresponding training data.The test data are arranged in time order in rows, and the data items are listed in Table 2. Abnormal data may exist in the raw data and need to be identified by the contestants themselves.
提交内容的数据格式及说明如表3所示，参赛者须对测试片段内车辆行驶里程进行补充，间隔符为英文逗号。
The data format and description of the submission are shown in Table 3, and the contestants are required to supplement the driving distance within test fragment with an English comma.
5.评分准则 Scoring criteria
根据提供的测试数据，计算预测值与实际结果的误差，公式如下：
Basedon the test data provided, the error between the prediction result and thereal-world value is calculated as follows:
电动汽车行驶SOC预测
1.竞赛题目 Question
电动汽车行驶SOC预测
Predictionof SOC in the driving process of electric vehicles
2.背景介绍 Background
由于电动汽车动力电池技术的限制，目前电动汽车的使用存在续驶里程不足，充电时间长，剩余续驶里程估计不准确等不便，这些缺陷在一定程度上阻碍了电动汽车的推广应用。电动汽车能耗的准确预测不仅能够帮助优化电动汽车控制策略从而减少能耗延长续驶里程，同时也能够为电动汽车剩余行驶里程的准确预测提供支持，进而为用户提供准确实用的驾驶辅助信息，帮助缓解里程焦虑。
Due tothe limitation of battery technology, electric vehicles have limited endurancemileage, long charging process and difficulty in remaining driving mileage prediction,which have become the main obstacles in the application of EVs. The accurateprediction of electric vehicle energy consumption can not only help optimize theelectric vehicle control strategy to reduce the energy consumption and extenddriving mileage, but also, it can provide support for the accurate predictionof the remaining driving mileage of electric vehicles, and provide drivers withaccurate and practical driving assistance information to help alleviate the mileageanxiety.
3.问题描述 Descriptionof the question
动力电池能耗受累计行驶里程、外界温度和道路情况等多因素耦合影响。参赛者要设计动力电池能耗预测模型，对行驶过程的动力电池能耗进行预测。在本题中，车辆在一段行驶中的能耗水平通过SOC变化表征，参赛者需要根据行驶片段的初始SOC以及行驶过程中的相关参数预测行驶片段内SOC的变化情况。
注：SOC以车辆车载终端上传数据为准。
Powerbattery energy consumption is affected by the multiple factors such ascumulative driving mileage, ambient temperature and road conditions. Contestantsare required to design a power battery energy consumption prediction model topredict the energy consumption of the battery during the driving process underreal-world driving condition. In this question, the energy consumption duringdriving process is indicated by the change of SOC, contestants are required touse the initial SOC and related parameters to predict the SOC change.
Note:The SOC is subject to the uploaded data of the vehicle terminal.
4.数据说明 Datadescription
本赛题数据来源于市面上某款电动汽车的实际运行数据，训练样本和测试样本数据由同一车型的不同车辆产生。
训练样本包含5辆电动汽车的实际运行数据（含行驶和充电等状态），数据按时间顺序排列成行，列名见表1。数据可能存在异常，需参赛者自行识别。
Thereal-world driving data of electric vehicles are used in this question, and thetraining data and test data are produced by different vehicles with the samemodel.
The training data consist of data of 5 vehicles (both driving state andcharging state are included), and the data are arranged in time order in rows.The data items are listed in Table 1. Abnormal data may exist in the raw dataand need to be identified by the contestants themselves.
测试样本由15辆车的150个行驶片段构成，每辆车10个行驶片段，测试样本中的0-4号车与训练样本中的0-4号车一一对应，5-14号车没有对应的训练样本数据。测试样本数据按时间顺序排列成行，列名见表2。数据可能存在异常，需参赛者自行识别。
Test data consist of 150 driving fragments of 15 vehicles, each vehicle provides 10 driving fragments. The vehicles No.0-4 in test data are corresponding to the vehicles No.0-4 in the training data, and vehicles No.5-14 have no corresponding training data.The test data are arranged in time order in rows, and the data items are listed in Table 2. Abnormal data may exist in the raw data and need to be identified by the contestants themselves.
注：测试片段的开始SOC是SOC发生跳变的第一帧，如（90，90，90，89，89,89）则选取SOC=89的第一帧作为开始，结束SOC是SOC发生跳变的最后一帧，如（23，23，23，22，22，22）则选取SOC=23的最后一帧作为结束。
Note: The starting SOC of test fragment is the first frame of SOC change, for example, (90, 90, 90, 89, 89, 89), the first frame of SOC=89 is selected as the start, and the end SOC is the last frame of SOC change, such as (23,23,23,22,22,22), the last frame of SOC=23 is selected as the ending.
提交内容的数据格式及说明如表3所示，参赛者须对测试区间结束时车辆SOC进行补充，间隔符为英文逗号。
The data format and description of the submission are shown in Table 3, and the contestants are required to supplement the vehicle SOC at the end of the test interval with an English comma.
5.评分准则 Scoringcriteria
根据提供的测试数据，计算预测值与实际结果的误差，公式如下：
Basedon the test data provided, the error between the prediction result and thereal-world value is calculated as follows:
训练数据集与测试数据集请至官网www.ncbdc.top页面【赛题与数据】-【创新组】查看。
点击【阅读原文】立即报名创新组竞赛！