Scholarlink

<p>In recent times, predicting the remaining useful life of lithium-ion batteries in electric vehicles has become increasingly significant in improving battery lifetime and vehicle performance. Traditional methods often lack accuracy and are more time-consuming. This research proposes a novel Adaptive Extremely Randomized Weight Adjustable Boosting-based Crossover Addax algorithm for accurate prediction of the remaining useful life of lithium-ion batteries in electric vehicles. In this research, an Extremely Randomized Trees and Adaptive Boosting with Weight-Adjustable Boosting model is deployed for predicting the lithium-ion battery’s remaining useful life. These ensemble methods dynamically adjust the weights of misclassified data points. Additionally, the Crossover Addax Optimization is applied to optimize the parameters of the ensemble model and reduce the overfitting issues, thereby reducing the computational burden. The effectiveness of the Adaptive Extremely Randomized Weight Adjustable Boosting-based Crossover Addax algorithm is validated using LG 18650HG2 Li-ion Battery data, the Hawaii Natural Energy Institute dataset, as well as the NMC111 dataset, and compared to existing remaining useful life prediction methodologies in terms of some common evaluation indicators. The experimental results demonstrate that the Adaptive Extremely Randomized Weight Adjustable Boosting-based Crossover Addax algorithm effectively predicted the lithium-ion battery’s remaining useful life in electric vehicles and achieved a high accuracy of 98.88%, a lower Root Mean Square Error of 2.763, and lower computation time of 0.6&#xa0;s compared to existing methodologies, underscoring its effectiveness in enhancing the performance of electric vehicles.</p>

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

Ensemble-Optimized Learning for Accurate Remaining Useful Life Estimation of Lithium-Ion Batteries in Electric Vehicles

  • 期刊论文

  • 发表日期：

    2025年7月24日

查看全文
  • J. Jeha,
  • M. Shunmuga Priyan

摘要

In recent times, predicting the remaining useful life of lithium-ion batteries in electric vehicles has become increasingly significant in improving battery lifetime and vehicle performance. Traditional methods often lack accuracy and are more time-consuming. This research proposes a novel Adaptive Extremely Randomized Weight Adjustable Boosting-based Crossover Addax algorithm for accurate prediction of the remaining useful life of lithium-ion batteries in electric vehicles. In this research, an Extremely Randomized Trees and Adaptive Boosting with Weight-Adjustable Boosting model is deployed for predicting the lithium-ion battery’s remaining useful life. These ensemble methods dynamically adjust the weights of misclassified data points. Additionally, the Crossover Addax Optimization is applied to optimize the parameters of the ensemble model and reduce the overfitting issues, thereby reducing the computational burden. The effectiveness of the Adaptive Extremely Randomized Weight Adjustable Boosting-based Crossover Addax algorithm is validated using LG 18650HG2 Li-ion Battery data, the Hawaii Natural Energy Institute dataset, as well as the NMC111 dataset, and compared to existing remaining useful life prediction methodologies in terms of some common evaluation indicators. The experimental results demonstrate that the Adaptive Extremely Randomized Weight Adjustable Boosting-based Crossover Addax algorithm effectively predicted the lithium-ion battery’s remaining useful life in electric vehicles and achieved a high accuracy of 98.88%, a lower Root Mean Square Error of 2.763, and lower computation time of 0.6 s compared to existing methodologies, underscoring its effectiveness in enhancing the performance of electric vehicles.