Research On Welding Shop Inverse Scheduling Methods Based On Improved Grey Wolf Optimizer

In the modern manufacturing industry,the welding process is a most critical technology of material forming and assembly,and the welding efficiency affects the efficiency of manufacturing.In the actual welding shop,many dynamic events may occur,such as machine breakdown,job with poor quality and urgent order add,which make the original scheduling scheme worse,resulting in the customers' need can't be satisfied.Meanwhile,welding is a high energy consumption process in the manufacturing industry.With the green manufacturing proposed in recent years,saving energy has become a priority.Welding shop inverse scheduling makes current scheduling scheme become optimal in the new situation,by minimum adjustment of processing parameters and coordination of relevant resources.Therefore,the research on welding shop inverse scheduling problem(WSISP)considering energy consumption has important theoretical value and practical significance.In this paper,the model and solution of WSISP are studied for the dynamic events according to actual welding shop characteristics,and minimizing the parameter adjustment and the total energy consumption are regarded as the optimal objectives.The single and multi-objectives WSISP models are established,and the Improved Grey Wolf Optimizer(IGWO)algorithm and Multi-objective Grey Wolf Optimizer(MOGWO)based on the original Grey Wolf Optimizer(GWO)is proposed,to solve the inverse scheduling problem,respectively.Finally,a case study considering a real-world welding shop is conducted to validate the proposed methods.Firstly,the WSISP is defined and the mathematical model is established according to the actual welding production characteristics and the inverse scheduling definition,and then the framework and solution are put forward,providing a theoretical basis for solving instance.Secondly,the model of single objective WSISP is established,and the inverse scheduling method based on IGWO algorithm is designed to solving the model.Matrix encoding method is applied,and variable neighborhood search strategy is used to improve the diversity of solutions.The results of simulation experiments prove the superiority of the proposed method for solving WSISP.Thirdly,a multi-objective mathematical model of WSISP is formulated,the MOGWO algorithm based on external archive and the selection of Pareto optimal solution strategy is proposed to solving the multi-objective problem.It is proved that the MOGWO algorithm is better than other algorithms through the simulation experiment.Then,a case study from an actual welding shop is conducted.The test result proves the validity of the inverse scheduling method proposed above for solving practical problem and satisfying the production requirement.Finally,a summary of the whole paper is made and several future study directions are prospected.