A swarm intelligence optimization algorithm for human-robot collaborative energy-efficient shop scheduling
With the intelligentization of manufacturing enterprises,the new development mode of"human-robot-cyber"comprehensive interconnection and integration has become the direction of the new generation of intelligent manufacturing.Collaborative manufacturing between workers and robots has become a common practice in modern manufacturing workshops,leading to increased challenges in resource allocation due to the departure from traditional,human-machine independent production scheduling mode.This complexity poses new challenges for algorithm design.Therefore,this paper addresses the characteristics of human-robot collaborative energy-efficient shop scheduling problem.The optimization objectives are to minimize the maximum completion time and total energy consumption.To achieve this,a swarm intelligence optimization algorithm is proposed.This algorithm features the following components:Firstly,a two-stage collaborative search framework is present.In the first stage,swarm intelligence optimization is employed to achieve rapid convergence and build an elite archive.In the second stage,a feedback collaborative search is conducted near elite solutions to explore a wide range of solutions.Secondly,multiple collaborative search operators are introduced to address human-robot resource allocation issues for rapid convergence.Lastly,in consideration of the problem's characteristics,problem-specific knowledge is extracted,and a multi-strategy collaborative initialization method is designed to provide a high-quality initial population.Additionally,multiple collaborative local search operations are proposed for efficient optimization of the two objectives.To validate the effectiveness of the proposed algorithm,15 cases of varying sizes are generated.The conducts parameter tuning,ablation,and comparative simulation experiments are adopted.The results demonstrate that the various collaborative methods proposed in this paper effectively enhance algorithm performance and search quality.Through comparisons with the latest relevant algorithms,the effectiveness of the proposed algorithm is evaluated,providing reliable guidance for scheduling and production in human-robot collaborative manufacturing.
万方数据
