AI-integrated self-powered active road stud-based energy harvesting for enhanced road safety in low-visibility

IF 9.2 2区工程技术 Q1 ENERGY & FUELS

Sustainable Materials and Technologies Pub Date : 2025-07-29 DOI:10.1016/j.susmat.2025.e01564

Touqeer Aslam , Ali Azam , Shoukat Ali Mugheri , Ammar Ahmed , Zutao Zhang , Mansour Abdelrahman , Juhuang Song , Chengliang Fan

{"title":"AI-integrated self-powered active road stud-based energy harvesting for enhanced road safety in low-visibility","authors":"Touqeer Aslam , Ali Azam , Shoukat Ali Mugheri , Ammar Ahmed , Zutao Zhang , Mansour Abdelrahman , Juhuang Song , Chengliang Fan","doi":"10.1016/j.susmat.2025.e01564","DOIUrl":null,"url":null,"abstract":"<div><div>This paper presents the experimental design and performance optimization of an AI-enabled Active Road Stud (ARS) system, utilizing a self-powered and self-balanced dual-slider crank mechanism to enhance safety and support smart infrastructure in low-visibility conditions. The proposed ARS harnesses kinetic energy from passing vehicles and converts it into electrical energy, offering a sustainable, off-grid solution for powering road lights, sensors, and traffic management tools. The study evaluates four road stud profiles: flat (S-1), parabolic (S-2), round (S-3), and wedge (S-4) to maximize energy conversion efficiency. Initially, numerical simulations were conducted in MATLAB/Simulink at varying frequencies corresponding to different vehicle speeds. These simulations revealed that the wedge-shaped (S-4) profile achieved the highest simulated performance, with a three-phase RMS power output of 34.71 W. The simulations were then validated through laboratory testing using a Mechanical Testing and Sensing (MTS) setup. Subsequently, real-time field experiments confirmed the practical performance of the S-4 profile, which achieved an RMS power output of 17.4 W and 9.21 V at a vehicle speed of 25 km/h and 3 Ω resistance, with an overall energy conversion efficiency of 66 %. Additionally, the ARS incorporates a deep learning-based condition monitoring system using a Long Short-Term Memory (LSTM) network to classify operational states (slow, fast, and failure) and predict maintenance needs. Real-time field tests also validated the system's reliability under adverse weather conditions such as fog, rain, and snow. The proposed ARS is a cost-effective, scalable solution for self-powered applications in both urban and rural environments.</div></div>","PeriodicalId":22097,"journal":{"name":"Sustainable Materials and Technologies","volume":"45 ","pages":"Article e01564"},"PeriodicalIF":9.2000,"publicationDate":"2025-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Sustainable Materials and Technologies","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S221499372500332X","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENERGY & FUELS","Score":null,"Total":0}

引用次数: 0

Abstract

This paper presents the experimental design and performance optimization of an AI-enabled Active Road Stud (ARS) system, utilizing a self-powered and self-balanced dual-slider crank mechanism to enhance safety and support smart infrastructure in low-visibility conditions. The proposed ARS harnesses kinetic energy from passing vehicles and converts it into electrical energy, offering a sustainable, off-grid solution for powering road lights, sensors, and traffic management tools. The study evaluates four road stud profiles: flat (S-1), parabolic (S-2), round (S-3), and wedge (S-4) to maximize energy conversion efficiency. Initially, numerical simulations were conducted in MATLAB/Simulink at varying frequencies corresponding to different vehicle speeds. These simulations revealed that the wedge-shaped (S-4) profile achieved the highest simulated performance, with a three-phase RMS power output of 34.71 W. The simulations were then validated through laboratory testing using a Mechanical Testing and Sensing (MTS) setup. Subsequently, real-time field experiments confirmed the practical performance of the S-4 profile, which achieved an RMS power output of 17.4 W and 9.21 V at a vehicle speed of 25 km/h and 3 Ω resistance, with an overall energy conversion efficiency of 66 %. Additionally, the ARS incorporates a deep learning-based condition monitoring system using a Long Short-Term Memory (LSTM) network to classify operational states (slow, fast, and failure) and predict maintenance needs. Real-time field tests also validated the system's reliability under adverse weather conditions such as fog, rain, and snow. The proposed ARS is a cost-effective, scalable solution for self-powered applications in both urban and rural environments.

查看原文本刊更多论文

基于人工智能的自供电主动道路能量收集，增强低能见度下的道路安全

本文介绍了一种基于人工智能的主动道路螺柱（ARS）系统的实验设计和性能优化，该系统利用自供电和自平衡双滑块曲柄机构来增强安全性，并支持低能见度条件下的智能基础设施。拟议中的自动驾驶系统利用过往车辆的动能并将其转化为电能，为道路照明、传感器和交通管理工具提供可持续的离网解决方案。该研究评估了四种道路螺柱形状：平面（S-1）、抛物线形（S-2）、圆形（S-3）和楔形（S-4），以最大限度地提高能量转换效率。首先在MATLAB/Simulink中对不同车速对应的不同频率进行数值模拟。仿真结果表明，楔形（S-4）型具有最高的仿真性能，三相均方根输出功率为34.71 W。然后通过使用机械测试和传感（MTS）设置的实验室测试验证了模拟。随后，实时现场实验验证了S-4型的实际性能，在车速为25 km/h、阻力为3 Ω时，S-4型的有效值输出功率为17.4 W、9.21 V，整体能量转换效率为66%。此外，ARS还集成了一个基于深度学习的状态监测系统，该系统使用长短期记忆（LSTM）网络对运行状态（慢速、快速和故障）进行分类，并预测维护需求。实时现场测试还验证了系统在恶劣天气条件下（如雾、雨、雪）的可靠性。提出的ARS是一种成本效益高、可扩展的解决方案，适用于城市和农村环境中的自供电应用。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Sustainable Materials and Technologies Energy-Renewable Energy, Sustainability and the Environment

CiteScore

13.40

自引率

4.20%

发文量

158

审稿时长

45 days

期刊介绍： Sustainable Materials and Technologies (SM&T), an international, cross-disciplinary, fully open access journal published by Elsevier, focuses on original full-length research articles and reviews. It covers applied or fundamental science of nano-, micro-, meso-, and macro-scale aspects of materials and technologies for sustainable development. SM&T gives special attention to contributions that bridge the knowledge gap between materials and system designs.