Scholarlink

<p>Designing a self-healing concrete that is going to be sustainable, self-sufficient in costs, and most importantly durable and strong throughout its desired lifecycle is the only solution to an ever-increasing complex set of infrastructure demands coupled with environmental constraints. These concrete mixture designs, involving complex, non-linear, multi-objective nature, often face optimization techniques of existing methods. Such traditional metaheuristics, though very useful, are not adaptable, slow in convergence, and not efficient in exploring large solution spaces under stringent performance constraints. This work presents a hybrid AI-quantum inspired multi-objective optimization framework for self-healing concrete design to deal with those challenges. The model integrates four developed computational techniques: (1) Quantum Inspired Differential Evolution with Adaptive Learning Mechanism (QIDE-ALM), improving exploration–exploitation balance using quantum bit-flipping and adaptive feedback; (2) Quantum-Accelerated Multi-Objective Particle Swarm Optimization (Q-MOPSO) that uses quantum tunneling to escape local optima and to speed up convergence; (3) Quantum-Driven Surrogate Modeling which uses quantum support vector machine and quantum neural network to reduce the computational burden on fast performance outcome prediction; and (4) Quantum Inspired Neural Networks for Multi-Objective Optimization (QINN-MO), dynamically learning complex relationships among mixture components by quantum Inspired weight modulation and architecture adaptations. Iterative implementation of this integrated model combines global searching with quick convergence and assessment, in conjunction with intelligent learning, generating Pareto-optimal concrete designs. The initial results show a tremendous improvement in performances: compressive strength of 50–55&#xa0;MPa, healing efficiency in the range of 90–95%, and lifecycle cost reduction of up to 20%. This framework is expected to prove potent, scalable, and computationally efficient in advancing concrete technology, thus entirely revolutionizing practices in civil infrastructure through intelligent process engineering of quantum-enhanced materials. </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

Hybrid quantum inspired multi-objective optimization framework for self-healing concrete using AI-driven metaheuristics

  • 期刊论文

  • 发表日期：

    2025年6月7日

查看全文
  • Aarti Karandikar,
  • Ashwini V. Zadgaonkar,
  • Rohit Pawar,
  • Ashwini C. Gote,
  • Tejas R. Patil,
  • Haytham F. Isleem

摘要

Designing a self-healing concrete that is going to be sustainable, self-sufficient in costs, and most importantly durable and strong throughout its desired lifecycle is the only solution to an ever-increasing complex set of infrastructure demands coupled with environmental constraints. These concrete mixture designs, involving complex, non-linear, multi-objective nature, often face optimization techniques of existing methods. Such traditional metaheuristics, though very useful, are not adaptable, slow in convergence, and not efficient in exploring large solution spaces under stringent performance constraints. This work presents a hybrid AI-quantum inspired multi-objective optimization framework for self-healing concrete design to deal with those challenges. The model integrates four developed computational techniques: (1) Quantum Inspired Differential Evolution with Adaptive Learning Mechanism (QIDE-ALM), improving exploration–exploitation balance using quantum bit-flipping and adaptive feedback; (2) Quantum-Accelerated Multi-Objective Particle Swarm Optimization (Q-MOPSO) that uses quantum tunneling to escape local optima and to speed up convergence; (3) Quantum-Driven Surrogate Modeling which uses quantum support vector machine and quantum neural network to reduce the computational burden on fast performance outcome prediction; and (4) Quantum Inspired Neural Networks for Multi-Objective Optimization (QINN-MO), dynamically learning complex relationships among mixture components by quantum Inspired weight modulation and architecture adaptations. Iterative implementation of this integrated model combines global searching with quick convergence and assessment, in conjunction with intelligent learning, generating Pareto-optimal concrete designs. The initial results show a tremendous improvement in performances: compressive strength of 50–55 MPa, healing efficiency in the range of 90–95%, and lifecycle cost reduction of up to 20%. This framework is expected to prove potent, scalable, and computationally efficient in advancing concrete technology, thus entirely revolutionizing practices in civil infrastructure through intelligent process engineering of quantum-enhanced materials.