Scholarlink

<p>Soft computing methodologies can be effectively employed with considerable accuracy for predicting the strength of asphalt mixes. This study introduces a comparative approach to assess the compressive strength of asphalt mixtures using adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN), and response surface methodology (RSM) within the framework of the Internet of Road Things (IoRT). IoRT plays a crucial role in this research by enabling real-time data collection on various parameters influencing polymer strength. Real-time sensor data from lamp posts is transmitted to a central IoRT platform. Multi-user fusion aggregates this data, enhancing reliability and addressing privacy concerns. This data fuels soft computing models (ANFIS, ANN, RSM) to predict asphalt strength in real-time. Additionally, the assessment of compressive strength through the response surface method, yielding a mean error of four, underscores the potential of this strategy for strength estimation. Accurately predicting the compressive strength of asphalt mixtures, particularly those incorporating diverse polymer reinforcements, presents a formidable challenge. RSM is applied in this research to estimate compressive strength under various conditions, such as modifying the ideal bitumen ratio, incorporating granular polymer-modified bitumen, and varying crushed particle ratios. The proposed concept introduces a polymer strength monitoring system utilizing road lamp posts as interconnected sensors and an Internet-of-Things platform as its central hub. This system delivers a transformative IoT-based solution, revolutionizing asphalt performance through enhanced strength, durability, and tailored characteristics. Data transmission between lamp posts and the back end via the Internet incorporates multi-user fusion concepts, generating authentic polymer strength conditions and addressing privacy concerns for industrial monitoring. ANN, ANFIS, and RSM accurately predicted asphalt strength, with RSM achieving the highest accuracy 99.8% and 91% MAPE, followed by ANFIS 97.1% and 88% MAPE and then ANN 94.5% and 85% MAPE. The accuracy of the response surface model is subsequently verified by comparing results with laboratory observations.</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

Soft Computing-Based Optimization Using ANFIS, ANN, and RSM to Predict the Strength of Polymer Adjusted Thin Layer Asphalt for Internet of Road Things (IoRT)

  • 期刊论文

  • 发表日期：

    2025年4月17日

查看全文
  • Sudipta Roy

摘要

Soft computing methodologies can be effectively employed with considerable accuracy for predicting the strength of asphalt mixes. This study introduces a comparative approach to assess the compressive strength of asphalt mixtures using adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN), and response surface methodology (RSM) within the framework of the Internet of Road Things (IoRT). IoRT plays a crucial role in this research by enabling real-time data collection on various parameters influencing polymer strength. Real-time sensor data from lamp posts is transmitted to a central IoRT platform. Multi-user fusion aggregates this data, enhancing reliability and addressing privacy concerns. This data fuels soft computing models (ANFIS, ANN, RSM) to predict asphalt strength in real-time. Additionally, the assessment of compressive strength through the response surface method, yielding a mean error of four, underscores the potential of this strategy for strength estimation. Accurately predicting the compressive strength of asphalt mixtures, particularly those incorporating diverse polymer reinforcements, presents a formidable challenge. RSM is applied in this research to estimate compressive strength under various conditions, such as modifying the ideal bitumen ratio, incorporating granular polymer-modified bitumen, and varying crushed particle ratios. The proposed concept introduces a polymer strength monitoring system utilizing road lamp posts as interconnected sensors and an Internet-of-Things platform as its central hub. This system delivers a transformative IoT-based solution, revolutionizing asphalt performance through enhanced strength, durability, and tailored characteristics. Data transmission between lamp posts and the back end via the Internet incorporates multi-user fusion concepts, generating authentic polymer strength conditions and addressing privacy concerns for industrial monitoring. ANN, ANFIS, and RSM accurately predicted asphalt strength, with RSM achieving the highest accuracy 99.8% and 91% MAPE, followed by ANFIS 97.1% and 88% MAPE and then ANN 94.5% and 85% MAPE. The accuracy of the response surface model is subsequently verified by comparing results with laboratory observations.