The invention relates to a new energy electric vehicle driving range prediction method, which is realized based on a machine learning algorithm, and is characterized in that the current running state of a vehicle and historical data in a past period of time are used as model input, and a vehicle driving range prediction result is used as model output; and generating a prediction model function based on a supervised learning method. The method specifically comprises the following steps of vehicle operation historical data information collection and arrangement, driving range label construction, data set feature screening, algorithm model construction, model iteration training optimization, model precision evaluation and model deployment real-time prediction. Compared with the prior art, the time sequence characteristic and non-time sequence characteristic information in the driving process of the new energy electric vehicle is fully excavated, and accurate prediction of the driving mileage is achieved. The risk that the vehicle breaks down due to the driving range prediction precision error problem is greatly reduced, and the method has the advantages of being high in timeliness, high in precision and high in robustness.
本发明涉及一种新能源电动汽车续驶里程预测方法,该方法基于机器学习算法实现,其特点是以车辆当前运行状态及过去一段时间内的历史数据作为模型输入,以车辆续驶里程预测结果作为模型输出,基于监督学习方法生成预测模型函数。具体包括以下步骤:车辆运行历史数据信息搜集整理、构造续驶里程标签、数据集特征筛选、算法模型构建、模型迭代训练优化、模型精度评估、模型部署实时预测。本发明与现有技术相比,充分挖掘了新能源电动汽车行驶过程中的时序特征与非时序特征信息,实现续驶里程的精确预测。大大减少了因续驶里程预测精度误差问题导致的车辆抛锚风险,具有时效性高、精准度高、鲁棒性强的特点。
New energy electric vehicle driving range prediction method
一种新能源电动汽车续驶里程预测方法
02.07.2024
Patent
Elektronische Ressource
Chinesisch
IPC: | B60L PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES , Antrieb von elektrisch angetriebenen Fahrzeugen / B60R Fahrzeuge, Fahrzeugausstattung oder Fahrzeugteile, soweit nicht anderweitig vorgesehen , VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Electric vehicle driving range prediction method and device
Europäisches Patentamt | 2020
|Energy-Optimal Driving Range Prediction for Electric Vehicles
British Library Conference Proceedings | 2017
|Electric vehicle electric quantity driving range evaluation method
Europäisches Patentamt | 2024
|Prediction Model of Driving Range of Electric Vehicle Based on Driver’s Driving Behavior
SAE Technical Papers | 2019
|