Scholarlink

Now a days electric vehicles (EV) are rapidly increasing on road due to their environmental and economic advantages. However to predict travel range of EVs remains a critical challenge due to several dynamic factors such as battery performance, road conditions and internal load. The road condition will affect the travel distance of EVs since more power required for upward road surface. In this paper, a machine learning (ML) approach using Linear Regression and Random Forest Regression is proposed to estimate the travel distance of an electric vehicle based on real time sensor inputs. A compact EV prototype was developed, equipped with voltage, current, temperature, GPS (Global Positioning System), temperature, humidity and, accelerometer and gyroscope sensors. Real time data was collected under various road conditions and used to train a ML model. The IoT (Internet of Things) and ML combined approach to demonstrated linear regression and random forest ML model utilized for predicting EV travel distance based on different road conditions. The model performance is evaluated on the test set using metrics such as R2 Score. The proposed framework offers a solution for users to predicting travel range of EV in different road conditions and to take necessary action accordingly.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

IoT-Enabled Machine Learning Approach for Predicting Battery Behavior in Electric Vehicles Under Varying Road Conditions

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Anand Burud,
  • Kamalakar Desai

摘要

Now a days electric vehicles (EV) are rapidly increasing on road due to their environmental and economic advantages. However to predict travel range of EVs remains a critical challenge due to several dynamic factors such as battery performance, road conditions and internal load. The road condition will affect the travel distance of EVs since more power required for upward road surface. In this paper, a machine learning (ML) approach using Linear Regression and Random Forest Regression is proposed to estimate the travel distance of an electric vehicle based on real time sensor inputs. A compact EV prototype was developed, equipped with voltage, current, temperature, GPS (Global Positioning System), temperature, humidity and, accelerometer and gyroscope sensors. Real time data was collected under various road conditions and used to train a ML model. The IoT (Internet of Things) and ML combined approach to demonstrated linear regression and random forest ML model utilized for predicting EV travel distance based on different road conditions. The model performance is evaluated on the test set using metrics such as R2 Score. The proposed framework offers a solution for users to predicting travel range of EV in different road conditions and to take necessary action accordingly.