使用改进约束处理方法的群体智能算法解决先进制造工艺问题
Solution to Advanced Manufacturing Process Problems using Cohort Intelligence Algorithm with Improved Constraint Handling Approaches
AI总结:
本文提出基于模量和双曲正切概率分布的两种新型约束处理方法,改进群体智能算法,用于优化水射流、磨料射流、超声和磨削等先进制造工艺,显著提升材料去除率,最高达127%。
摘要:
近年来,各种基于人工智能(AI)的优化元启发式算法被提出并应用于多种问题。群体智能(CI)算法是一种受社会启发的优化技术,已成功应用于解决设计、制造、供应链、医疗保健等多个领域的多个无约束和约束现实世界问题。通常,现实世界问题本质上是受约束的。尽管大多数进化算法(EAs)能有效解决无约束问题,但当涉及约束时,其性能会退化。本文提出了两种基于模量和双曲正切概率分布的新型约束处理方法。提出了带有基于三角、模量和双曲正切的约束处理方法的约束CI算法,并将其应用于优化先进制造工艺,如水射流加工(WJM)、磨料射流加工(AJM)、超声加工(USM)和磨削工艺。使用所提出的CI算法获得的解决方案与当代算法如遗传算法、模拟退火、教学学习优化等进行了比较。所提出的方法在满足硬约束的情况下实现了材料去除率2%-127%的最大化。与GA相比,带有双曲正切概率分布的CI在AJMB、AJMD、WJM、USM和磨削工艺中分别实现了MRR的15%、2%、2%、127%和4%的改进,有助于提高生产率。本文的贡献为所提出的约束处理方法在解决复杂约束问题中的进一步适用性开辟了多条途径。
英文摘要:
Recently, various Artificial Intelligence (AI) based optimization metaheuristics are proposed and applied for a variety of problems. Cohort Intelligence (CI) algorithm is a socio inspired optimization technique which is successfully applied for solving several unconstrained & constrained real-world problems from the domains such as design, manufacturing, supply chain, healthcare, etc. Generally, real-world problems are constrained in nature. Even though most of the Evolutionary Algorithms (EAs) can efficiently solve unconstrained problems, their performance degenerates when the constraints are involved. In this paper, two novel constraint handling approaches based on modulus and hyperbolic tangent probability distributions are proposed. Constrained CI algorithm with constraint handling approaches based on triangular, modulus and hyperbolic tangent is presented and applied for optimizing advanced manufacturing processes such as Water Jet Machining (WJM), Abrasive Jet Machining (AJM), Ultrasonic Machining (USM) and Grinding process. The solutions obtained using proposed CI algorithm are compared with contemporary algorithms such as Genetic Algorithm, Simulated Annealing, Teaching Learning Based Optimization, etc. The proposed approaches achieved 2%-127% maximization of material removal rate satisfying hard constraints. As compared to the GA, CI with Hyperbolic tangent probability distribution achieved 15%, 2%, 2%, 127%, and 4% improvement in MRR for AJMB, AJMD, WJM, USM, and Grinding processes, respectively contributing to the productivity improvement. The contributions in this paper have opened several avenues for further applicability of the proposed constraint handling approaches for solving complex constrained problems.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。