The Methods Of Robotic Manufacturing Cell Integrated Process Planning And Scheduling

Process planning and job shop scheduling are independent modules in manufacturing cell,which are intrinsically linked and mutually restricted.The different interactions of the two modules directly affect the processing capacity,device utilization and production efficiency.In traditional manufacturing enterprises,process planning and job shop scheduling are assigned to independent departments.According to this mode,the planned process route may become invalid in actual operation.The most essential reason of this phenomenon is that the transfer operation is carried out manually in traditional enterprises,which brings a lot of uncertainties.Considering that the industrial robot has the characteristics of certainty and re-programmability,and with the application of robot technology in industrial production.The research on robotic manufacturing cell integrated process planning and scheduling(RMC-IPPS)has important theoretical value and practical significance.The essential research works are as follows:(1)The multi-objective RMC-IPPS solving strategy of optimization before decision is determined.In the optimization stage,the non-inferior solutions of RMC-IPPS are solved by multi-objective evolutionary algorithm.In the decision stage,the satisfactory solution is selected from the non-inferior solutions,which based on the analytical hierarchy process.Under this strategies,this paper designs a high-dimensional multi-objective evolutionary algorithm based on NSGA-Ⅲ algorithm,clustering analysis and simulated annealing algorithm(cluster-NSGA-Ⅲ).The algorithm makes up for the design defect that NSGA-Ⅲ algorithm does not have local search.Combined with the characteristics of unsupervised learning of clustering analysis,it mines the internal relationship between individuals in the population,making the algorithm take into account both global search and local search.(2)Aiming at the various flexibility involved in the integrated process planning and scheduling(IPPS),a multi-level linear coding method was proposed.This method can effectively express the solution of the IPPS with a simple data structure,and realize the overall search of the IPPS.The benchmark examples are used to verify the proposed method,and the improved algorithm is compared with the algorithm in the literature to verify the feasibility and effectiveness of the proposed method in solving the IPPS.(3)On the basis of the established IPPS optimization method,a robot scheduling strategy based on greedy idea is designed to solve the problem of blank/semi-finished product transfer in the cell.The strategy selects the current optimal state at each step,so that the production cycle will not be delayed too much due to the existence of the handling process.A practical production example is given to illustrate the proposed method.The proposed method is tested and compared with NSGA-Ⅲ algorithm and the preference based multi-objective particle swarm optimization algorithm.The test results verify the superiority of the proposed method.(4)Based on the fuzzy set theory,the uncertain RMC-IPPS models are established.The models not only consider the uncertainty factors in the operation of traditional machines,but also consider the certainty of robot operation.The decoding of code method,the practical production application example is improved to meet the test standard of uncertain RMC-IPPS,and the algorithm is used to solve the calculation example,which verifies the effectiveness of the method.