Prediction Method of Charging Energy for Power Battery of Electric Vehicle
-
摘要: 提出了一种基于机器学习中Stacking模型的电池充电能量预测方法,该方法通过对充电数据进行数据探索、特征工程和模型筛选,选取RMSE作为预测结果的评价指标,最后采用Stacking模型对充电能量作出预测。为了验证Stacking模型的预测结果,将Stacking模型与采用单个算法模型的预测结果进行对比,以确保方案的可行性。其结果表明,采用该模型进行预测时,其预测结果的RMSE分值为0.104 1,实现了比单个算法模型更好的预测效果。
-
关键词:
- 电动汽车 /
- 充电能量 /
- Stacking模型 /
- 机器学习
Abstract: In the past, the impact of battery performance evaluation and electric vehicles on the grid has often neglected the impact of battery charging energy, and the charging energy of the battery is also an important indicator in the charging process. To this end, a prediction method of electric charging energy based on the stacking model for machine learning is proposed. The method performs data processing, feature extraction and model screening on the data of charging, and selects the RMSE as the evaluation index of the prediction result. Finally, the stacking model is used to predict the charging energy. In order to verify the prediction results of the Stacking model, the stacking model is compared with the prediction results with a single algorithm model to ensure the feasibility of the scheme. The results show that with the present model for prediction, the RMSE score of the prediction result is 0.104 1, which achieves better prediction than that via single algorithm model.-
Key words:
- electric vehicle /
- charging energy /
- Stacking model /
- machine learning
-
表 1 数据分类与说明
类型 特征字符 数据类型 说明 时间信息类 Charge_start_time int 充电开始时间 Charge_end_time int 充电结束时间 车辆信息类 Vehicle_id string 车辆唯一标志码 Mileage float 充电开始时刻车辆仪表里程/km 电池信息类 Charge_start_soc float 充电开始时刻动力电池soc/% Charge_end_soc float 充电结束时刻动力电池soc/% Charge_start_U float 充电开始时刻动力电池总电压/V Charge_end_U float 充电结束时刻动力电池总电压/V Charge_start_I float 充电开始时刻动力电池总电流/A Charge_end_I float 充电结束时刻动力电池总电流/A Charge_max_temp float 充电过程中电池系统最大温度/℃ Charge_min_temp float 充电过程中电池系统最小温度/℃ Charge_energy float 此充电过程的充电能量/ (kW·h) 表 2 动力电池充电能量的基本特征
类型 特征字符 数据类型 说明 时间信息类 Charge_time int 充电时间 Day_of_week int 充电时间日标记 Weekend int 充电时间周末标记 Charge_start_hour int 充电开始时标记 Hour_of_day int 充电时间日夜标记 电池信息类 Soc_time float 充电过程中电池SOC的增长率 Charge_diff_U float 充电起止电压差值/V Charge_diff_I float 充电起止电流差值/A Charge_diff_temp float 充电起止温度差/℃ Charge_aver_U float 充电过程动力电池平均总电压/V Charge_aver_I float 充电过程动力电池平均总电流/A Charge_start_power float 充电开始功率/W Charge_end_power float 充电结束功率/W Charge_aver_power float 充充电过程中平均功率/W -
[1] Joshi K, Lakum A. Assessing the impact of plug-in hybrid electric vehicles on distribution network operations using time-series distribution power flow analysis[C]//Proceedings of 2014 IEEE International Conference on Power Electronics, Drives and Energy Systems (PEDES). Mumbai, India: IEEE, 2014 [2] Habib S, Kamran M, Rashid U. Impact analysis of vehicle-to-grid technology and charging strategies of electric vehicles on distribution networks–a review[J]. Journal of Power Sources, 2015, 277: 205-214 doi: 10.1016/j.jpowsour.2014.12.020 [3] Leou R C, Su C L, Lu C N. Stochastic analyses of electric vehicle charging impacts on distribution network[J]. IEEE Transactions on Power Systems, 2014, 29(3): 1055-1063 doi: 10.1109/TPWRS.2013.2291556 [4] Qian K J, Zhou C K, Allan M, et al. Modeling of load demand due to EV battery charging in distribution systems[J]. IEEE Transactions on Power Systems, 2011, 26(2): 802-810 doi: 10.1109/TPWRS.2010.2057456 [5] Arias M B, Bae S. Electric vehicle charging demand forecasting model based on big data technologies[J]. Applied Energy, 2016, 183: 327-339 doi: 10.1016/j.apenergy.2016.08.080 [6] Arias M B, Kim M, Bae S. Prediction of electric vehicle charging-power demand in realistic urban traffic networks[J]. Applied Energy, 2017, 195: 738-753 doi: 10.1016/j.apenergy.2017.02.021 [7] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016Zhou Z H. Machine learning[M]. Beijing: Tsinghua University Press, 2016 (in Chinese) [8] Tang X P, Liu B Y, Lv Z, et al. Observer based battery SOC estimation: using multi-gain-switching approach[J]. Applied Energy, 2017, 204: 1275-1283 doi: 10.1016/j.apenergy.2017.03.079 [9] 朱嘉欣, 包雨恬, 黎朝. 数据离群值的检验及处理方法讨论[J]. 大学化学, 2018, 33(8): 58-65 doi: 10.3866/PKU.DXHX201802008Zhu J X, Bao Y T, Li Z. Discussion on the method for testing and treating outliers[J]. University Chemistry, 2018, 33(8): 58-65 (in Chinese) doi: 10.3866/PKU.DXHX201802008 [10] Castillo-Gutiérrez S, Lozano-Aguilera E D, Estudillo-Martínez M D. A new proposal to adjust a straight line to a normal Q-Q plot[J]. Journal of Mathematics and System Science, 2012, 2(5): 327-333 [11] Taylor N. Realised variance forecasting under Box-Cox transformations[J]. International Journal of Forecasting, 2017, 33(4): 770-785 doi: 10.1016/j.ijforecast.2017.04.001 [12] 吴刘仓, 黄丽, 戴琳. Box-Cox变换下联合均值与方差模型的极大似然估计[J]. 统计与信息论坛, 2012, 27(5): 3-8 doi: 10.3969/j.issn.1007-3116.2012.05.001Wu L C, Huang L, Dai L. Maximum likelihood estimation for joint mean and variance models of Box-Cox transformation[J]. Statistics & Information Forum, 2012, 27(5): 3-8 (in Chinese) doi: 10.3969/j.issn.1007-3116.2012.05.001 [13] 孙吉贵, 刘杰, 赵连宇. 聚类算法研究[J]. 软件学报, 2008, 19(1): 48-61 doi: 10.3724/SP.J.1001.2008.00048Sun J G, Liu J, Zhao L Y. Clustering algorithms research[J]. Journal of Software, 2008, 19(1): 48-61 (in Chinese) doi: 10.3724/SP.J.1001.2008.00048 [14] Wong T T. Performance evaluation of classification algorithms by k-fold and leave-one-out cross validation[J]. Pattern Recognition, 2015, 48(9): 2839-2846 doi: 10.1016/j.patcog.2015.03.009 [15] Rodriguez J D, Perez A, Lozano J A. Sensitivity analysis of k-fold cross validation in prediction error estimation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(3): 569-575 doi: 10.1109/TPAMI.2009.187 [16] Mentaschi L, Besio G, Cassola F, et al. Problems in RMSE-based wave model validations[J]. Ocean Modelling, 2013, 72: 53-58 doi: 10.1016/j.ocemod.2013.08.003 [17] Zheng H T, Yuan J B, Chen L. Short-term load forecasting using EMD-LSTM neural networks with a xgboost algorithm for feature importance evaluation[J]. Energies, 2017, 10(8): 1168 doi: 10.3390/en10081168 [18] Ma X J, Sha J L, Wang D H, et al. Study on a prediction of P2P network loan default based on the machine learning LightGBM and XGboost algorithms according to different high dimensional data cleaning[J]. Electronic Commerce Research and Applications, 2018, 31: 24-39 doi: 10.1016/j.elerap.2018.08.002 [19] Wang D H, Zhang Y, Zhao Y. LightGBM: an effective miRNA classification method in breast cancer patients[C]//Proceedings of 2017 International Conference on Computational Biology and Bioinformatics. New Jersey, USA: ICCBB, 2017: 7-11 [20] Zhang M, Fei X, Liu Z H. Short-term traffic flow prediction based on combination model of xgboost-lightgbm[C]//Proceedings of 2018 International Conference on Sensor Networks and Signal Processing (SNSP). Xi′an: IEEE, 2018 [21] Pavlyshenko B. Using stacking approaches for machine learning models[C]//Proceedings of 2018 IEEE Second International Conference on Data Stream Mining & Processing. Lviv, Ukraine: IEEE, 2018